Independent Public Service Pensions Commission: Final Report

Independent Public ServicePensions Commission

Independent Public Service Pensions Commission:Final Report

10 March 2011

Independent Public ServicePensions Commission

Independent Public Service Pensions Commission:Final Report

March 2011

Official versions of this document are printed on 100% recycled paper. When you have finished with it please recycle it again.

If using an electronic version of the document, please consider the environment and only print the pages which you need and recycle them when you have finished.

© Crown copyright 2011

You may re-use this information (not including logos) free of charge in any format or medium, under the terms of the Open Government Licence. To view this licence, visit http://www.nationalarchives.gov.uk/doc/open-government-licence/ or write to the Information Policy Team, The National Archives, Kew, London TW9 4DU, or e-mail: [email protected].

ISBN 978-1-84532-855-9PU1128

http://www.nationalarchives.gov.uk/doc/open-government-licence/

http://www.nationalarchives.gov.uk/doc/open-government-licence/

mailto:psi%40nationalarchives.gsi.gov.uk?subject=

Independent Public Service Pensions Commission: Final Report - 10 March 2011

Contents

Contents ................................................................................................................... 1Foreword by Lord Hutton of Furness ........................................................................ 3Executive Summary .................................................................................................. 7

Chapter 1: The case for reform revisited ................................................................. 21Chapter 2: The deal ................................................................................................ 37Chapter 3: The design ............................................................................................ 49Chapter 4: The controls .......................................................................................... 87Chapter 5: Applying the design ............................................................................ 103Chapter 6: A transparent and effective system ...................................................... 121Chapter 7: Delivering the change ..........................................................................147Chapter 8: Conclusions .........................................................................................157

Annex a: List of recommendations .........................................................................159Annex b: Case studies of best practice ................................................................... 163Annex c: Technical modelling ................................................................................171Annex d: Summary of evidence ............................................................................. 187Annex e: List of charts, diagrams and tables ...........................................................195Annex f: Glossary ................................................................................................. 199Annex g: Abbreviations ......................................................................................... 205Annex h: Bibliography .......................................................................................... 207

3


Foreword by Lord Hutton of Furness

In my interim report last October, I set out the case for further reforms to public service pensions. I explained the principles that I believed should govern any future reforms and the values that should underpin this process going forward.

In this, my final report I am setting out how I believe it is possible for public service employees to continue to have access for the foreseeable future, to good quality, sustainable and fairer defined benefit pension schemes. For this to happen there will need to be comprehensive reform – reforms that can balance the legitimate concerns of taxpayers about the present and future cost of pension commitments in the public sector as well as the wider need to ensure decent levels of retirement income for millions of people who have devoted their working lives in the service of the public. I am taking as given the new landscape in which the Consumer Prices Index is the measure of inflation and a rise in employee contributions is imminent.

Establishing a relationship of trust and confidence going forward will be very important. Ministers have already accepted the conclusions of my interim report that pension reform must not simply become a race to the bottom. This has helped set the right parameters for the debate about the next steps. Ministers have also been clear that the pension promises that have been made must be honoured. This is important too in order to correct the widespread view that pension reform invariably means losing pension rights that have already been accrued. It does not. The more this can be stressed the better. Trade unions for their part have also been willing to accept the need in the past for changes to public service pensions that address the shifting sands of economic, demographic and social change and as a consequence there have already been some significant reductions in the projected cost of public service pensions.

Effective reform will require this dialogue to continue. I strongly believe that a process must be established in which both sides to the debate can have full access to information, can properly communicate their views, and understand the nature and purpose of any changes. Establishing such a process will be fundamental to building consensus over long term pension reform in the public sector; how people are treated in this process will be as important as the changes to pension schemes themselves. This must not however become a recipe for procrastination. We need to get on with the process of change if we are to maximise the benefits from reform. These benefits include certainty about the future, a fairer distribution of the enormous risks and costs involved in maintaining any form of defined benefit pension, and long term financial sustainability.

In considering which reforms to propose I have considered a wide range of options from notional and collective defined contribution, lump sum accrual or cash balance mechanisms

4


and career average defined benefit schemes. Each has much to commend them. They are all capable to different extents of achieving the four principles of reform I set out in the interim report. I have however been influenced by the need for simplicity, clarity and ease of implementation. The more complicated a scheme is to describe, the harder it will be to implement and the less likelihood there is of active engagement and support.

My main recommendation therefore is that Government should replace the existing final salary pension schemes with a new career average scheme and, when everything is ready, move existing members to the new scheme for future accruals. I am recommending this course of action as I believe this is the fairest way of spreading the effect of change across the generations, and represents the quickest way of ending the in-built bias against those public service employees whose pay stays low over their career, inherent in final salary schemes. Maintaining the link to final salary for the purposes of calculating the value of a person’s accrued rights under the existing schemes will however ensure fair treatment for those who have built up rights in these schemes and will mean that those closest to retirement, perhaps in their 50s today, who have less time to adjust are least affected and all existing scheme members retain the link to final salary for the years they have already accrued.

As I set out in my interim report rising life expectancy has led to a substantial increase in the proportion of adult life that a public service worker can expect to spend in retirement. To adjust to this change I am recommending that Normal Pension Age is linked to State Pension Age and tracks planned increases. In principle the link to State Pension Age would apply across all public service workers, as this marks the end of a working life that may span professions and sectors. However, for the uniformed services – the armed forces, police and firefighters – where pension age has historically been lower to reflect the unique nature of their work a pension age of 60 is appropriate.

This link to State Pension Age will address rising longevity, the main risk to the sustainability of public service pensions. But to manage other pressures I also recommend that ministers should set a clear cost ceiling for these new schemes going forward – I suggest the percentage of pensionable pay paid by the taxpayer – with automatic stabilisers built into their design to keep future costs under more effective control. These stabilisers will mean that scheme members might need to increase their contributions, or take a smaller pension, the choice should be the subject of discussion with staff but an automatic default must be agreed. This will allow a new framework of public service pensions to be established which will be much more resilient and better able to absorb the kind of shocks that have profoundly affected the viability of defined benefit schemes over the last few decades. It will also provide confidence to taxpayers that there will be firm limits set on how much they can be expected to contribute to public service pensions.

I also consider that there is a powerful case for more independent oversight and much stronger governance of all the public service pension schemes. This should keep government, taxpayers and scheme members better informed about the financial health of these schemes. There should be minimum standards set for scheme administration. There is also a proper and legitimate role for representatives of the workforce to be formally involved in these new governance arrangements.

5


The current legal framework for public service pensions needs a complete overhaul and I am making specific recommendations in this area.

As a package, these changes will make public service pension schemes simpler and more transparent, fairer to those on low and moderate earnings, better able to deal with the changes that we know are coming to our economy and our society and will therefore help ensure greater prospects for sustainability over the long term. Ministers and ultimately Parliament itself must decide how generous they want public service pensions to be. For my part, I consider that there is a strong argument for designing future public service pension schemes in such a way that they at least deliver the replacement rates set out in the Turner Commission in 2005. Making these changes will not be easy or straightforward. Confronting the fundamental challenges posed to our pension system caused by rising life expectancy, managing inflation and in some cases investment risks over the long term, ensuring productivity in the wider economy and value for money for taxpayers requires us to make difficult choices.

In overcoming these challenges we will all be required to think differently about our working lives as well as our retirement. But it does not mean we have to give up the idea of designing a pension system that can deliver decent retirement incomes. If the changes I am recommending are implemented as a whole, then I see no reason why this should not be possible. I remain convinced that this is well within our grasp. My recommendations are therefore designed to make the changes that will be necessary to ensure that this vital part of our national retirement savings system is to continue to both set high standards – something the public service should, in my view always seek to do – and command the trust and confidence of all those concerned.

I believe this is a balanced deal that will ensure public service workers continue to have good pensions and taxpayers can have confidence that the costs are controlled. This deal is set out in full in the report and summarised in the diagram below.

Over the last nine months I have been fortunate to have had the support and assistance of an extraordinarily talented group of people within the Commission. I want to thank each of them for the help they have given me. I would also like to thank my panel of experts (Ron Amy, Professor Nicholas Barr, Lord Michael Bichard, Professor David Blake, Niki Cleal, Baroness Jeannie Drake, Carl Emmerson, Professor John Hills and Professor Alasdair Smith) for their comments and contributions. What appears in the pages that follow are of course my conclusions alone. But I strongly believe they hold out the prospect of managing rather than succumbing to the powerful forces of change that are requiring individuals, companies and Governments all over the world to re-think their approach to pensions.

Lord Hutton of Furness

6


TH

E D

EAL

Publ

ic se

rvic

e w

orke

rsTa

xpay

ers

A g

ood

pens

ion

in re

tire

men

t: a

leve

l of p

ensio

n th

at a

t lea

st m

eets

agre

ed a

dequ

ate

stand

ards

of

pens

ion

- tak

en to

geth

er w

ith fu

ll sta

te p

ensio

n th

is sh

ould

del

iver

on

ave

rage

mor

e th

an tw

o th

irds o

f pre

-retir

emen

t sal

ary

for t

hose

be

low

med

ian

inco

me.

A d

efi n

ed b

enefi

t p

ensi

on:

a pe

nsio

n ba

sed

on a

vera

ge sa

lary

inde

xed

by a

vera

ge e

arni

ngs o

ver

your

car

eer.

The

desig

n sh

ould

ben

efi t

the

maj

ority

of m

embe

rs w

ho

do n

ot h

ave

the

high

sala

ry g

row

th re

war

ded

in a

fi na

l sal

ary

sche

me.

Acc

rued

rig

hts p

rote

cted

: th

e ye

ars y

ou h

ave

alre

ady

wor

ked

prov

ide

a pe

nsio

n at

you

r cur

rent

pe

nsio

n ag

e lin

ked

to y

our fi

nal

sala

ry. T

his w

ill p

rote

ct e

xisti

ng

staff

from

full

impa

ct o

f cha

nge

in p

ropo

rtion

to th

eir a

ge a

nd c

aree

r le

ngth

.

Fair

pro

cess

of c

hang

e:

the

deta

ils o

f cha

nge

shou

ld b

e th

e su

bjec

t of c

onsu

ltatio

n w

ith st

aff

and

unio

ns.

Bet

ter

man

agem

ent o

f sch

emes

:im

prov

ed st

anda

rds o

f gov

erna

nce

and

adm

inist

ratio

n w

ith st

aff

invo

lvem

ent.

Fair

er sh

arin

g of

ben

efi t

of l

ivin

g lo

nger

: pu

blic

serv

ice

wor

kers

will

ove

r tim

e be

exp

ecte

d to

wor

k lo

nger

- m

ost t

o sta

te p

ensio

n ag

e - b

efor

e th

ey ta

ke th

eir p

ensio

n. T

his

will

reba

lanc

e th

e pr

opor

tion

of a

dult

life

spen

t in

retir

emen

t.

Futu

re-p

roof

ed:

pens

ion

age

in m

ost p

ublic

serv

ice

sche

mes

will

be

expe

cted

to

keep

in li

ne w

ith c

hang

es to

life

exp

ecta

ncy

thro

ugh

a lin

k to

sta

te p

ensio

n ag

e ch

ange

s.

Fixe

d co

st:

the

Gov

ernm

ent s

houl

d es

tabl

ish a

fi xe

d co

st fo

r the

em

ploy

ers’

cont

ribut

ion

to p

ublic

serv

ice

pens

ion

sche

mes

. If c

ost g

row

s be

yond

this

leve

l act

ion

will

be

take

n to

get

bac

k to

this

leve

l.

Gre

ater

tran

spar

ency

of c

ost:

fi gur

es fo

r the

cur

rent

and

futu

re e

xpec

ted

cost

of p

ublic

serv

ice

pens

ions

shou

ld b

e pu

blish

ed m

ore

regu

larly

, con

siste

ntly

and

tra

nspa

rent

ly.

Sing

le le

gal f

ram

ewor

k:

publ

ic se

rvic

e pe

nsio

ns sh

ould

hav

e a

new

lega

l fra

mew

ork

with

co

nsist

ent a

ppro

ach

to c

ontro

l and

gov

erna

nce.

7


Executive Summary

The case for reform revisited

Ex.1 In its interim report the Independent Public Service Pensions Commission found that the current public service pensions structure has been unable to respond flexibly to workforce and demographic changes in the past few decades. This has led to:

• rising value of benefits due to increasing longevity;

• unequal treatment of members within the same profession;

• unfair sharing of costs between the employee, the employer and taxpayers; and

• barriers to increasing the range of providers of public services.

Ex.2 The interim report recommended long-term structural reform as the issues could not, in the Commission’s view, be dealt with through traditional final salary defined benefit schemes. But neither could they be dealt with appropriately through a funded, individual account, defined contribution model for all employees. The Government accepted the report’s conclusions and affirmed its commitment to maintaining some form of defined benefit pension provision for public service employees.

Ex.3 The Commission takes as given the recent changes to public service pension schemes, including the use of the Consumer Prices Index as the measure of inflation and an imminent rise in employee contributions. These changes have reduced cost pressures, but have not addressed fundamental longer-term structural problems.

Ex.4 Since the interim report the wider pensions landscape has continued to evolve. Of particular relevance to the Commission’s work is the 2011 Pensions Bill, which proposes accelerating the planned increases in the State Pension Age. The State Pension Age is now planned to be 66 for both men and women by 2020.

Ex.5 When considering the possible reform options for public service pensions the Commission used its framework of principles outlined in the interim report. Public service pensions should, in the Commission’s view, be:

• affordable and sustainable;

• adequate and fair;

8


• supporting productivity; and

• transparent and simple.

Ex.6 In the light of these principles, the Commission has concluded that public service pension schemes should aim to ensure adequate incomes in retirement for those who have worked a full career in the public service. Pensions are also an important part of the reward package and can therefore help public service employers to recruit and retain staff. But they are an inflexible tool for workforce management; a task in which they are unlikely to provide value for money.

Recommendation 1: The Government should make clear its assessment of the role of public service pension schemes. Based on its framework of principles, the Commission believes that the primary purpose is to ensure adequate levels of retirement income for public service pensioners.

Recommendation 2: Pensions will continue to be an important element of remuneration. The Commission recommends that public service employers take greater account of public service pensions when constructing remuneration packages and designing workforce strategies. The Government should make clear in its remits for pay review bodies that they should consider how public service pensions affect total reward when making pay recommendations.

The deal

Ex.7 The package of reforms recommended by the Commission is a balanced deal that will deliver fair outcomes for public service workers and for taxpayers and build trust and confidence in the system. Public service workers should receive a good pension in retirement and their accrued rights must be protected. They must also be involved in the process of change and they have a right to expect schemes to be well-run with greater transparency. And taxpayers must be able to feel confident that risks and costs are shared fairly: in particular that the cost of increasing longevity is being managed and that there are safety valves in place to control future cost. There also needs to be independent assurance on the sustainability of public service pensions. The deal set out by the Commission is designed to meet these objectives.

Ex.8 In the interim report the Commission set out its view that public service pensions must deliver adequate pension levels. In the final report we have defined this further to provide a clear benchmark for the minimum level of benefit required.

9


Recommendation 3: The Government should ensure that public service schemes, along with a full state pension, deliver at least adequate levels of income (as defined by the Turner Commission benchmark replacement rates) for scheme members who work full careers in public service. Employers should seek to maximise participation in the schemes where this is appropriate. Adequate incomes and good participation rates are particularly important below median income levels.

Ex.9 Protecting accrued rights is a prerequisite for reform both to build trust and confidence and to protect current workers from a sudden change in their pension benefits or pension age. It is also right that those closest to retirement will be least affected by any changes to scheme design.

Recommendation 4: The Government must honour in full the pension promises that have been accrued by scheme members: their accrued rights. In doing so, the Commission recommends maintaining the final salary link for past service for current members.

Ex.10 The main risks within defined benefit schemes are: investment; inflation; salary; and longevity risk. While government, as a large employer, is capable of bearing the majority of the risk associated with pension saving efficiently, and should continue to do so through a defined benefit pension, present schemes involve too much risk for government and the taxpayer.

Ex.11 There should be a fairer sharing of risk between government (and ultimately taxpayers) and scheme members than exists within the present schemes. Achieving this will mean moving current members to new schemes. Allowing current members to continue to accrue further benefits in the present schemes for many decades would be unfair and inequitable to the new members coming behind them.

Recommendation 5: As soon as practical, members of the current defined benefit public service pension schemes should be moved to the new schemes for future service, but the Government should continue to provide a form of defined benefit pension as the core design.

Ex.12 But the taxpayer should also have confidence that public service pension costs are under control and are sustainable. That requires mechanisms in the scheme design to share cost and risk fairly and a fixed cost ceiling to assure cost control.

Ex.13 Transparency and effective oversight of public service schemes is required for public service workers and taxpayers to have confidence in the system and improve the quality of debate about the future of public service pensions. Currently there is inconsistency in what scheme data and assessments, such as valuations, are published and such information is often difficult to access. This lack of transparency prevents comparisons and hinders analysis.

10


Recommendation 6: All public service pension schemes should regularly publish data which, as far as possible, is produced to common standards and methodologies and is then collated centrally. This information should be of a quality that allows simple comparisons to be made across Government, between schemes and between individual Local Government Pension Scheme (LGPS) Funds.

The design

Ex.14 The Commission’s view is that defined benefit should continue to be the core design for public service pensions as an efficient design for a large employer to share risk with employees. But as set out in the interim report, and expanded further in this report, final salary does not provide the right design for future public service schemes. Final salary schemes unfairly benefit high flyers who can receive up to twice as much in pension payments per £100 of contributions. It exposes taxpayers to salary risk (the risk that higher than expected salary rises increase the cost of providing pensions), which should be borne by the scheme member who benefits from the salary rise. And final salary creates a barrier to employees moving from the public to private sector. These inherent problems of final salary schemes impact on fairness and sustainability and have led the Commission to conclude that an alternative model should be chosen for the future.

Ex.15 Career average schemes allow pension to be accrued on the basis of earnings in each year of service. In these schemes future earnings do not affect past years’ pension accrual so mobility between sectors is easier, salary risk remains with members and the unfairness of big benefits to high flyers is removed. Career average benefits can be delivered through a cash balance scheme. Cash balance could provide greater flexibility to alter benefits in the light of changes in longevity.

Ex.16 So both career average and cash balance schemes could provide a good match against the Commission’s principles and in terms of the distribution of risks between member and taxpayer. On balance, the Commission has decided to recommend career average as the option that provides more certainty for members, is better understood and will be more practical to implement. The Commission is not recommending specific levels for accrual rates, indexation and employee contributions as these determine cost, which is a matter for the Government. The Government will need to make a decision about these parameters after consultation with scheme members.

Recommendation 7: A new career average revalued earnings (CARE) scheme should be adopted for general use in the public service schemes.

Ex.17 In a career average scheme the level of indexation and the accrual rate determines how different types of members are impacted by the scheme. The Commission favours indexation by average earnings during the accrual phase to maintain the value of the benefits,

11


offset by lower accrual rates for a given cost. Uprating benefits in line with average wage increases ensures that benefit accrual in early years is broadly linked to earnings increases and therefore better relates to a member’s level of earnings at retirement. It is a better indexation measure than prices which unfairly benefits the years worked late in a career compared to the years worked at the start and is therefore unfair between younger and older members.

Recommendation 8: Pension benefits should be uprated in line with average earnings during the accrual phase for active scheme members. Post-retirement, pensions in payment should be indexed in line with prices to maintain their purchasing power and adequacy during retirement.

Ex.18 Regarding the indexation of deferred members’ benefits, there is a trade-off to be made. If the indexation measure were the same as for active members this would favour mobility. If it were lower, for example, if active members’ benefits were indexed by earnings and deferred members’ benefits by prices, this would favour retention. The Government should decide on whether pre-retirement indexation for deferred members is on an earnings based measure or prices based measure, as this decision will need to be based on the explicit objectives that government has about recruitment and retention versus mobility.

Ex.19 Employee contribution rates across public service pensions schemes vary considerably, both in level and structure, as the result of historic developments and negotiations over a number of years at scheme-specific level. Contribution rates should be set so that members appreciate the value of their pensions but not so high that they lead to scheme members, especially at lower income levels, opting out of the pension scheme. Any transition to a more uniform structure for employee contribution rates will be complex and must be considered in the context of overall remuneration.

Ex.20 Higher earning individuals are better positioned than lower earners to bear some of the risks associated with pension provision. In addition, higher earners are likely to live longer than lower earners and so will derive more benefit from their scheme for the same level of pension. It is desirable for the design of public service pension schemes to recognise this.

Ex.21 However, the introduction of a cap on pensionable earnings or hybrid schemes (schemes which have both defined benefit and defined contribution features) does not seem attractive due to the complexity this introduces to the system, the significant transitional issue of the cash flow loss to government revenue and the Commission’s view that there are advantages to having senior management and their staff in the same scheme. However, there is a case for tiered contribution rates to reflect the different characteristics of higher earners.

12


Recommendation 9: A single benefit design should apply across the whole income range. The differing characteristics of higher and lower earners should be addressed through tiered contribution rates. The Government should consider the trade off between affordability and the impact of opt outs on adequacy when setting member contribution levels.

Ex.22 More generally, if state pension benefits were to change in future the Government would need to consider the future benefit and contribution arrangements in public service pensions to ensure they are still fair.

Ex.23 Choice in pension schemes can have significant advantages for scheme members and government alike. However, choice should be limited within the core design of the schemes, because of the evidence of undersaving when choice is available and the need to avoid undue complexity. But members should be given choice over their ability to make additional contributions to their pensions in a simple and transparent manner: this will encourage greater provision for retirement. Choice over when to draw pension benefits could be facilitated by providing information on how retirement income would change with the age at which the pension was taken, with information starting to be provided perhaps 5 or 10 years before Normal Pension Age. Actuarial enhancement and reduction should be applied in the new schemes in response to late or early retirement, with caps on pension accrual either increased or lifted entirely.

Ex.24 At present, most public service pension schemes provide for abatement, where some people who return to work in a job covered by the same scheme from which they draw their pension receive a reduced pension as a result. This can have a significant negative impact on an employee’s effective salary if he returns to work.

Recommendation 10: Members should have greater choice over when to start drawing their pension benefits, so they can choose to retire earlier or later than their Normal Pension Age and their pension would be adjusted accordingly on an actuarially fair basis. Flexible retirement should be encouraged and abatement of pensions in its current form for those who return to work after drawing their pensions should be eliminated. In addition, caps on pension accrual should be removed or significantly lifted.

Ex.25 Replacing or removing abatement of pensions and increasing or removing caps on pension accrual would have cost implications for pension schemes which will need to be considered as part of the wider package.

The controls

Ex.26 Whichever overall design for public service schemes is adopted, longevity risk is the principal risk that needs to be managed. Life expectancy has increased dramatically in the last few decades and future changes are uncertain. These changes have been recognised within

13


the State Pension system. Following that lead would be an appropriate way to help members bear pre-retirement longevity risk and will increase the chances of scheme members having adequate retirement incomes.

Recommendation 11: The Government should increase the member’s Normal Pension Age in the new schemes so that it is in line with their State Pension Age. The link between the State Pension Age and Normal Pension Age should be regularly reviewed, to make sure it is still appropriate, with a preference for keeping the two pension ages linked.

Ex.27 The introduction of the link to the State Pension Age, which will initially move Normal Pension Ages to 65, will move the proportion of adult life in retirement for public service pension scheme members back to about a third: roughly where it was in the 1980s. The current State Pension Age of 65 is already the Normal Pension Age for most new entrants to public service pension schemes. Moving to this for future accrual will more fairly distribute the benefits between scheme members. In the long term, the timetabled increases in State Pension Age should help to keep the proportion of adult life in retirement for members around this level, on current life expectancy projections.

Ex.28 This measure and the other design features proposed by the Commission should achieve much of the Commission’s aim regarding sharing risks and costs fairly between employees and the Government. However, an additional safety valve, a cost ceiling based on the proportion of their total pensionable pay bill, is needed in case costs within the new schemes increase due to factors not taken account of in the scheme design. This will ensure that public service pensions remain affordable and sustainable.

Recommendation 12: The Government, on behalf of the taxpayer, should set out a fixed cost ceiling: the proportion of pensionable pay that they will contribute, on average, to employees’ pensions over the long term. If this is exceeded then there should be a consultation process to bring costs back within the ceiling, with an automatic default change if agreement cannot be reached.

Ex.29 What is included within this cost ceiling is a matter for the Government to determine in consultation with employees and their representatives. However, if the ceiling is exceeded measures will need to be taken to bring costs back down below it. There should be a default stabilising mechanism that could take the form of an increase in employee contributions or a decrease in accrual rates which would automatically reduce costs if negotiations between employers and scheme members were unsuccessful.

14


Applying the design

Ex.30 There are differences between the individual public service schemes in factors such as the distribution of pensionable pay, average career length and life expectancy. But, for most schemes, these differences are generally similar to, or smaller than, those seen within the schemes. The modernised schemes generally apply standard features to their members and this is accepted as an appropriate approach, as tailoring schemes to all the differences seen would create a complex and costly system and would be likely to reduce member understanding. This would seem to suggest that for most schemes variation in pension features is not the most appropriate way to deal with these differences where they are seen.

Recommendation 13: The Commission is not proposing a single public service pension scheme, but over time public service pensions should move towards a common framework for scheme design as set out in this report. However, in some cases, for example, the uniformed services, there may need to be limited adaptations to this framework.

Ex.31 The uniformed services (the armed forces, police and firefighters) are in a somewhat different position, given that the pension ages in the uniformed services schemes still generally reflect an assumption that pension for the majority of long-serving members should be payable from age 55 or less. This assumption may no longer match expectations, given the increases in life expectancy that have been seen since the 19th and first half of the 20th century when these pensions ages were set.

Ex.32 But this does not take away from the fact that the nature of the work the uniformed services perform is unique and that this needs to be reflected in their Normal Pension Ages. The modernised firefighters scheme has struck a balance between recognising these changes in life expectancy, but also recognising the unique nature of the service provided by scheme members. The Commission’s view is that the Normal Pension Age in this scheme, 60, should be seen as setting a benchmark for the uniformed services as a whole. This position will need to be kept under regular review to make sure it is still appropriate, given future changes in life expectancy projections and experience of healthy life expectancy.

Recommendation 14: The key design features contained in this report should apply to all public service pension schemes. The exception is in the case of the uniformed services where the Normal Pension Age should be set to reflect the unique characteristics of the work involved. The Government should therefore consider setting a new Normal Pension Age of 60 across the uniformed services, where the Normal Pension Age is currently below this level in these schemes, and keep this under regular review.

Ex.33 The LGPS provides a set of final salary-based benefits similar to those in many of the unfunded schemes. The membership also shares characteristics with membership of very large unfunded schemes and there are overlaps in coverage with unfunded schemes. There

15


may be a higher proportion of part-time, lower-earning members in the LGPS, but there are also many such members in schemes like the Principal Civil Service Pension Scheme and the National Health Service Pension Scheme.

Ex.34 There are good reasons for Government to fund pensions from future taxation and finance them on an unfunded basis if this is done in a transparent way and taken account of within fiscal planning in an appropriate manner. The Commission has therefore concluded that the funding boundary, with the LGPS funded and other major schemes unfunded, should remain where it is. This will continue the mixed approach to the funding of public service pension schemes.

Recommendation 15: The common design features laid out in this report should also apply to the LGPS. However, it remains appropriate for the Government to maintain the different financing arrangements for the LGPS in future, so the LGPS remains funded and the other major schemes remain unfunded.

Ex.35 As for the categories of people who should in future be entitled to join public service pension schemes, it is ultimately for the Government to decide how much pensions risk it is willing to bear in order to meet its wider policy objectives. While continuing access to public service pension schemes helps to remove the pensions barrier for external contractors, there are good reasons for the Government to limit access, including the increased long term risk government would bear in relation to those schemes.

Ex.36 In addition, since the publication of the Commission’s interim report the Government has announced reviews of the Fair Deal policy and the discount rate, which are relevant to many of the issues around access to schemes. The Commission expects that the outcome of these reviews would, at least in part, help to facilitate the Government’s aim for increased plurality of provision for public services. And a move to a new public service pension scheme design, as laid out in this report, should also help to remove some of the barriers to plurality of service provision.

Recommendation 16: It is in principle undesirable for future non-public service workers to have access to public service pension schemes, given the increased long-term risk this places on the Government and taxpayers.

A transparent and effective system

Ex.37 Currently in the public service pension schemes there is not always a clear separation of duties between those responsible for the governance of public service pension schemes and those delivering the benefits to scheme members. This can lead to a lack of transparency and of clarity as to who is responsible for what. And members of public service pension schemes

16


are sometimes not formally represented in the existing governance arrangements, which can lead to a lack of adequate member involvement in analysis and decisions.

Recommendation 17: Every public service pension scheme (and individual LGPS Fund) should have a properly constituted, trained and competent Pension Board, with member nominees, responsible for meeting good standards of governance including effective and efficient administration. There should also be a pension policy group for each scheme at national level for considering major changes to scheme rules.

Ex.38 Not all public service pension schemes communicate with members on a regular basis. Currently it is a requirement of defined contribution schemes in the private sector that they provide members with an annual benefit statement: this is not the case for defined benefit schemes (the great majority of public service pension schemes), which only have to provide a statement if requested. Yet the provision of information supports the general requirement to improve awareness of pensions and to assist members taking ownership of their pension requirements.

Recommendation 18: All public service pension schemes should issue regular benefit statements to active scheme members, at least annually and without being requested and promote the use of information technology for providing information to members and employers.

Ex.39 Public service pension schemes are not subject to external independent regulation in the way that private sector schemes are. While this full regulatory system would not be appropriate for public service schemes it seems reasonable that members of all pension schemes should be clear about how their scheme is run and by whom and be confident that their interests are being protected.

Recommendation 19: Governance and the availability and transparency of information would be improved by government establishing a framework that ensures independent oversight of the governance, administration and data transparency of public service pension schemes. Government should consider which body or bodies, including, for example, The Pensions Regulator, is most suitable to undertake this role.

Ex.40 While the recommendations in this report on scheme features aim to create a sustainable public service pensions system for the future, it is important that given the implications for long-term public finances this is monitored by an independent body to make sure it is achieved. Given its role to provide independent advice on the public finances the Office for Budget Responsibility is best placed to play this role.

17


Recommendation 20: When assessing the long term sustainability of the public finances, the Office for Budget Responsibility should provide a regular published analysis of the long term fiscal impact of the main public service pension schemes (including the funded LGPS).

Ex.41 Management of investment funds is an additional aspect of pension scheme governance for the funded public service pension schemes, most notably for the LGPS. While there are many areas of good practice around the management of these funds there are also areas that could be improved around publication of comparable Fund data and assessment of sustainability.

Recommendation 21: Centrally collated comprehensive data, covering all LGPS Funds, should be published including Fund comparisons, which, for example, clarify and compare key assumptions about investment growth and differences in deficit recovery plans.

Ex.42 Good administration is a key enabler in the delivery of accurate and timely pension payments. Currently a clear definition of what good administration (and governance) looks like in public service pension schemes is not readily available.

Recommendation 22: Government should set what good standards of administration should consist of in the public service pension schemes based on independent expert advice. The Pensions Regulator might have a role, building on its objective to promote good administration. A benchmarking exercise should then be conducted across all the schemes to assist in the raising of standards where appropriate.

Ex.43 A number of commentators have also suggested that public service pension schemes offer scope for the streamlining their administration functions and more shared arrangements. This is relevant to all schemes, but particularly for the administration of the locally run schemes. A number of local authorities have already begun to explore opportunities to share administrative services and contracts and the Commission’s view is that this should be encouraged.

Recommendation 23: Central and local government should closely monitor the benefits associated with the current co-operative projects within the LGPS, with a view to encouraging the extension of this approach, if appropriate, across all local authorities. Government should also examine closely the potential for the unfunded public service schemes to realise greater efficiencies in the administration of pensions by sharing contracts and combining support services, including considering outsourcing.

18


Ex.44 There is a complex and inconsistent overall legal architecture that has developed piecemeal over the last 100 years or more. There is a strong case for introducing new, overarching, primary legislation to set the new public service pension scheme framework. This would provide greater transparency, simplicity and certainty that the reforms would satisfy common basic principles.

Recommendation 24: The Government should introduce primary legislation to adopt a new common UK legal framework for public service schemes.

Delivering the change

Ex.45 Implementation is key: the first stage will require the detailed development of proposals by the Government including the Government’s decision on affordability issues while applying the common set of design principles.

Recommendation 25: The consultation process itself should be centrally co-ordinated: to set the cost ceilings and timetables for consultation and overall implementation. However, the consultation on details should be conducted scheme by scheme involving employees and their representatives.

Ex.46 This will avoid a cumbersome process while permitting any necessary variation in timetables between schemes. It will also provide for standardisation while allowing individual schemes to enter into negotiations. Each consultation, as required by law, will need to be accompanied by a full equality impact assessment to allow for deeper consideration of issues the reforms may pose for various groups. And there should be early upfront communication with scheme members, to encourage their participation in the consultation process.

Ex.47 The Commission recognises that these reforms cannot be achieved overnight. There are several steps that will need to be taken before the necessary legislative process can be started and the consequent administrative changes made. And these steps are crucial in ensuring that the reforms are a success and deliver sustainable public service pension schemes for the foreseeable future.

Recommendation 26: The Commission’s view is that even allowing for the necessary processes it should be possible to introduce the new schemes before the end of this Parliament and we would encourage the Government to aim for implementation within this timeframe.

Ex.48 Meeting the implementation timetable would be facilitated by considering delivery issues, particularly scheme administration, at an early stage. And there needs to be a clear

19


governance structure for the implementation plan. This is important for several reasons. First, it increases transparency in the process. Second, it allows the monitoring of delivery against targets and gives early warning if it looks like the timetable is slipping. Third, it allows individuals responsible for delivering the reforms to be held to account.

Recommendation 27: Best practice governance arrangements should be followed for both business as usual and the transformation process, for each scheme. And there will also need to be the right resource, on top of business as usual, to drive the reforms; particularly given the challenging timescale and scope of the reforms.

Ex.49 An important part of delivering this change and managing the transition is protecting existing members. Members will of course have their past service rights protected. The Commission has also noted in the evidence submitted to it the widespread expectations among public servants that the final salary link would be maintained. The Commission is sympathetic to the argument that this would be in line with the principle of accrued rights. And given the major transition involved in moving to the reformed schemes it would also be fair.

Ex.50 The Commission’s expectation is that existing members who are currently in their 50s should, by and large, experience fairly limited change to the benefit which they would otherwise have expected to accrue by the time they reach their current scheme normal pension age. This would particularly be the case if the final salary link is protected for past service, as the Commission recommends.

Ex.51 The reforms proposed by the Commission should, as a whole, deliver a framework for public service pensions that is affordable and sustainable, adequate and fair, supports productivity and is transparent and simple. It represents a balanced deal for public service workers and taxpayers that should deliver a long-lasting settlement in which they can have trust and confidence.

20


21


1 The case for reform revisited

Box 1.A: Summary

• In the interim report, the Commission outlined the case for long-term structural reform of public service pension schemes and concluded that the final salary structure of most current schemes can be unfair to employees and taxpayers. In the context of uncertain and increasing longevity, the Commission also felt that current scheme designs are not sufficiently robust to ensure the sustainability of public service pensions.

• The Commission felt that there was a rationale for short-term cost savings in recognition of the substantial unanticipated increases in longevity. In practice these savings could only be realised by increasing member contributions. The Commission recommended that any increase should be managed so as to protect the low paid, and if possible staged.

• The Government accepted the report’s conclusions and committed to maintaining some form of defined benefit (DB) pensions provision for public service employees. It announced staggered and progressive increases in member contributions, of three percentage points on average. In response to the report, the Government also announced public consultations on the discount rate and on the Fair Deal policy that protects the pension entitlements of employees who are transferred out of the public sector.

• The Government should make clear its assessment of the role of public service pension schemes. Based on its framework of principles, the Commission believes that the primary purpose is to ensure adequate levels of retirement income for public service pensioners.

• Pensions will continue to be an important element of remuneration. The Commission recommends that public service employers take greater account of public service pensions when constructing remuneration packages and designing workforce strategies. The Government should make clear in its remits for pay review bodies that they should consider how public service pensions affect total reward when making pay recommendations.

The interim report

1.1 The Commission published its interim report on 7 October 2010. This showed that the life expectancy of those reaching the age of 60 has increased by an unprecedented amount since the Second World War (Chart 1.A).

22


Chart 1.A: Period life expectancies for those reaching age 60 – general population

60

65

70

75

80

85

90

Life

exp

ecta

ncy

(yea

rs)

Year

Males Females

Source: Life tables, Office for National Statistics.

Note: Period life expectancies represent the amount of time an individual is expected to live under the assumption that

mortality rates are equal to the experience of other individuals in that year (see Box 4.C of the interim report). When

life expectancy is increasing, period life expectancies tend to understate somewhat the actual anticipated lifespan of a

particular cohort reaching the age of 60.

1.2 These improvements have consistently been underestimated (see Chart 4.B). As a result, time in retirement, and thus pension costs, have been much higher than originally expected. A female pensioner in the NHS scheme who retired at the age of 60 in 2010 could expect to spend around 45 per cent of their adult life in retirement, compared to around 30 per cent for pensioners in the 1950s (see Chart 4.A). These developments are discussed in more detail in Chapter 4.

1.3 There have been significant reforms to the main public service pension schemes over the last decade, including increased pension ages for new members and a change in the indexation of pensions from RPI to CPI indexation. Some of these changes have reduced projected benefit payments in the coming decades. For the interim report the Commission asked the Government Actuary’s Department (GAD) to project future public service pensions expenditure. It projected benefit payments to fall gradually to around 1.4 per cent of GDP in 2059-60, after peaking at 1.9 per cent of gross domestic product (GDP) in 2010-11.

1.4 Future costs are inherently uncertain and sensitive to assumptions on life expectancy, size of workforce, earnings growth and the implementation of reforms. Chart 1.B demonstrates the possible impact of altering some of these assumptions. Given the current

23


design of public service pension schemes, the general public cannot be sure that schemes will remain sustainable in the future.

Chart 1.B: Projected benefit payments as a percentage of GDP – sensitivity analysis

0.0%

0.2%

0.4%

0.6%

0.8%

1.0%

1.2%

1.4%

1.6%

1.8%

2.0%

1999/2000 2009/10 2019/20 2029/30 2039/40 2049/50 2059/60

Perc

enta

ge o

f pro

ject

ed G

DP

YearSource: GAD projections for IPSPC and IPSPC analysis.

Note: The fan chart shows how the projections would be affected by altering assumptions about productivity growth,

public service workforce growth and life expectancy (see Annex C of the interim report).

1.5 The interim report also assessed the fairness of public service pensions. It showed that the final salary design of most current schemes is much more beneficial to high flyers than to those with slower salary growth. High flyers can receive almost twice as much in pension payments per £100 of employee contribution than low flyers.1

1.6 New analysis by the Commission, using a dataset of over one million members of the Local Government Pension Scheme (LGPS), confirms that this is an important issue. Though data are not available to track the difference between high and low flyers directly, Chart 1.C shows that those who retire on a higher salary in a final salary pension scheme receive a significantly higher annual pension per £100 of employee contributions. The median annual pension payout for an employee who retired on a salary in the highest fifth was £42 for each £100 of contributions. This is almost 30 per cent higher than the pension payout for someone who retired on a salary in the lowest fifth. Larger benefits for high flyers, relative to their contributions, are an inherent feature of final salary pension schemes, but this does not happen in several other possible scheme designs.2

1 IPSPC (2010), Interim Report, Box 5.C, p.94.

2 Pensions Policy Institute (2010), The future of the public sector pensions, Chapter 5.

24


Chart 1.C: Median annual pension receipts for each £100 of contributions

05

1015202530354045

1st quintile (earning up to

£12,160)

2nd quintile (earning up to

£15,282)

3rd quintile (earning up to

£18,808)

4th quintile (earning up to

£25,449)

5th quintile (earning over

£25,449)

Med

ian

annu

al p

ensi

on re

ceip

ts fo

r ea

ch £

100

of c

ontr

ibut

ions

(£)

Source: IPSPC analysis of scheme data returns.

Note: Based on current pensioners who retired from the Local Government Pension Scheme in 2008.

1.7 The results are especially striking because the LGPS has long had a system of differentiated contributions. Historically, manual workers paid less than non-manual employees, while contributions in the reformed scheme are tiered on the basis of income, with higher earners paying more. Analysis by the Pensions Policy Institute confirmed that the current tiered contributions in the NHS scheme are insufficient to offset the higher benefits received by high flyers.3

1.8 Another potential unfairness arises from the concentration of previous reforms in most schemes on new members. This implies that new joiners often receive significantly less generous pension entitlements than their colleagues in pre-reform schemes. For instance, the Pensions Policy Institute estimated that, on average, post-reform public service pension schemes now provide benefits worth around 18 per cent of salary, compared to 20 per cent of salary for members of the pre-reform schemes (Table 1.A).4 For a median earner, this is equivalent to a reduction in pension benefits of about £560 per annum.5 If the planned increases in member contributions (described below) are enacted, the average effective benefit rate for members of reformed schemes could be expected to fall to around 15 per cent of salary.

3 IPSPC (2010), Interim Report, p.95.

4 IPSPC (2010), Interim Report, p.97.

5 Based on a full-time public sector employee – see Office for National Statistics (2010), ‘The changing face of public

sector employment 1999-2009,’ Economic and Labour Market Review, July.

25


Table 1.A: Recent changes in average effective employee benefit rates for the main schemes

Average employee benefit rates Members of old schemes – % of pay

Members of new schemes – % of pay

with RPI indexation 24 21

with CPI indexation (from April 2011) 20 18

Source: Pensions Policy Institute.

Note: Based on the seven main public service pension schemes (NHS, teachers, LGPS, civil service, police, fire and

armed forces).

1.9 The Commission determined that current public service final salary pension schemes could harm productivity. By leaving almost all risks with employers, they can make it difficult to attract new providers to achieve gains in the efficiency and quality of services. The ‘Fair Deal’ policy requires private and voluntary sector employers to provide comparable pensions when they take on public sector workforces. Smaller private and voluntary sector employers are often unwilling to take on such risks. Labour market flexibility could also be undermined by final salary pension schemes, since they create strong barriers to mobility from the public to the private sector.

1.10 The interim report showed that pension scheme membership in the private sector has declined significantly in the last fifteen years, with fewer than 35 per cent of private sector employees now members of employer-sponsored pension schemes (Chart 1.D). Pension membership has remained high in the public sector, and the Commission made clear that any structural reforms to public service pensions should aim to maintain or increase levels of employee participation. This is discussed further in Chapter 3.

26


Chart 1.D: Membership of employer-sponsored pension schemes among UK employees

0

20

40

60

80

100

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

Per c

ent

Year

DB scheme DC scheme

Unknown No provision

Public sector pension provision

Source: Annual Survey of Hours and Earnings.

Note: There was a methodological change between 2005 and 2006, which has a marginal impact on results. The

‘Unknown’ category was introduced in 2005.

Interim conclusions

1.11 The Commission firmly rejected the claim that current public service pensions are ‘gold plated.’ The average pension paid to pensioner members is around £7,800 per year, while the median payment is around £5,600. It also rejected the idea of a ‘race to the bottom,’ emphasising instead that public service pensions should ensure adequate retirement incomes for those who have devoted their careers in the service of the wider community.

1.12 The Commission determined that longer term structural reform of pensions was required, because current schemes had proved unable to respond flexibly to changes in working lives and longevity. They had also resulted in an unfair balance of risks between scheme members and taxpayers. The inherent problems of final salary pension schemes, particularly in terms of fairness and sustainability, led the Commission to decide that alternative models should be chosen for the future (Box 1.B).

27


Box 1.B: How do current final salary schemes measure up to the Commission’s principles?

• Affordable and sustainable: Due to the link between pension benefits and final earnings, the majority of salary risk (the risk that higher than expected salary rises increase the cost of providing pensions) in current schemes is borne by the Government. Members receive a substantial increase in pension rights from salary rises as they approach retirement, with the costs falling on public service employers.

• Adequate and fair: Any pension design is capable of delivering adequate pensions. The available evidence suggests that current final salary schemes in the public sector do achieve this. However, high flyers typically derive more value from final salary schemes than low flyers, leading to unfairness between scheme members. A high-flying employee could receive almost twice as much in pension payments per £100 of contributions than a low flyer. In addition, the balance of risks between the government and the member is one-sided, leading to unfairness between the taxpayer and scheme members.

• Supporting productivity: Final salary schemes restrict labour market mobility. Depending on the Government’s underlying objectives for the scheme this may be desirable for the retention of skilled higher earners. However, at the macroeconomic level a more flexible labour market should increase efficiency across the economy as a whole.

• Transparent and simple: The current schemes are reasonably well understood and simple to administer. However, transparency is an issue since it can be difficult to ascertain the benefit derived from the scheme, relative to contributions paid in, for higher and lower earners, and high and low flyers. For the taxpayer, there is little transparency of expected cost, since this depends on future pay developments.

1.13 The Commission was also asked to consider the case for short-term savings. It concluded that increased longevity meant that there was a clear rationale for short-term savings, to ensure a fairer distribution of costs between taxpayers and members. Because most benefit payments in the short term reflect accrued rights (see Chart 2.B), only an increase in member contributions could produce significant immediate savings. The Commission recommended that any increase in contributions should be managed so as to protect the low paid, and staged in order to minimise the impact on opt out rates. The Commission felt that there was also a case for reviewing the discount rate6 used to set contribution rates in the unfunded public service pension schemes, and for reviewing the ‘Fair Deal’ policy, which determines the pension rights of public sector workers transferred to the private sector.

The Government’s response

1.14 The Government responded to the Commission’s interim report in the Spending Review published on 20 October 2010.7 The Government accepted the conclusions of

6 For a definition of the discount rate see IPSPC (2010), Interim Report, Box 4.D.

7 HM Treasury (2010), Spending Review 2010.

28


the report and committed to continuing with a form of defined benefit pension for public service workers.

1.15 It also agreed to carry out a public consultation on the discount rate; this was published on the 9 December 2010 and closed on 3 March 2011. In addition, on 3 March 2011 the Government launched a consultation on the Fair Deal policy.

1.16 The Government further announced that it would implement progressive changes to the level of employee contributions, leading to an additional saving of £1.8 billion a year by 2014-15, to be phased in from April 2012.8 This is equivalent to a rise in contribution rates of three percentage points on average. The armed forces will be exempt from the increases, as recommended by the Commission.

Updating the landscape

1.17 The economic landscape, along with that of pensions, is continually changing. The Office for Budget Responsibility (OBR) published its Economic and fiscal outlook in November 2010. This forecast GDP growth to rise steadily from 2010, peaking at 2.9 per cent in 2013, before easing slightly as a result of demographic changes. Annual public service pensions expenditure is expected to rise to £33.2 billion by 2015-16, an increase from £32.9 billion forecast in June 2010.

1.18 The Commission’s preferred measure of the cost of public service pensions is the level of benefit payments as a percentage of GDP. This can give a good sense of the share of national income that has to be devoted to public service pensions expenditure. Other criteria can also be useful in certain contexts, though the widely-used net cash expenditure figure (the gap between current contributions received and current benefit payments) is not an appropriate measure. As well as being inherently volatile, it is a mismatch between contributions made in respect of future benefits and payments of previously accrued benefits, and so provides no insight into long-term affordability.9 As a proportion of national income, public service pensions spending is expected to be broadly stable in the short term, remaining between 1.7 and 1.8 per cent of GDP throughout the forecast period (Chart 1.E).

8 The figure for expected savings is in addition to the £1 billion of savings already expected through cap and share

mechanisms. It does not include any savings from the LGPS.

9 The National Audit Office writes that this measure ‘has no advantages as a measure of pension costs’ (National Audit

Office (2010), The impact of the 2007-08 changes to public service pensions).

29


Chart 1.E: Projected public service pensions expenditure

1.50%

1.55%

1.60%

1.65%

1.70%

1.75%

1.80%

1.85%

1.90%

1.95%

2.00%

2010-11 2011-12 2012-13 2013-14 2014-15 2015-16

Per c

ent o

f pro

ject

ed G

DP

Year

Source: Economic and fiscal outlook, Office for Budget Responsibility, Tables 4.2 and 4.18, 2010.

1.19 The OBR’s revised forecasts of GDP growth, inflation and the public sector workforce could also have an impact on the longer-term cost of public service pensions. Relative to its Budget forecasts, the OBR expects a somewhat smaller reduction in the number of public sector workers (falling to 5.17 million in 2014-15) and a somewhat smaller gap between RPI and CPI inflation. Both of these factors could be expected to increase long-term public service pension spending, but they are partially offset by expectations of slightly higher GDP growth in the medium term. Overall, the Commission assesses that the changed projections suggest a modest increase in long-term public service pension expenditure as a proportion of GDP, but this does not imply any substantial alteration to the conclusions reached in the interim report, as shown in Chart 1.B.

1.20 In the wider public sector, remuneration practices could be affected by any changes resulting from Will Hutton’s review of public sector pay, discussed in Box 1.C.

30


Box 1.C: The Hutton Review of Fair Pay in the Public Sector

The Hutton Review of Fair Pay, chaired by Will Hutton, published its interim report in December 2010. It reported that pay dispersion has increased in the public sector, largely following private sector trends (though to a much lesser extent). Though top pay often appears to be below that of private sector comparators, there are developments in governance, transparency and competitiveness that could enhance the perceived fairness of pay awards. There is also a good case for referring pay to a pay multiple.

The central conclusions of the report were that:

• private sector top pay has raced ahead – and in some areas of the public sector it has been felt that executive pay needs to try to keep up. Looked at in terms of responsibilities, public sector executive pay often looks well behind that in the private sector. For instance, the Permanent Secretary of the Home Office (earning about £200,000) is responsible for a budget of over £10 billion. Private sector chief executives with budgets of that size have median salaries in excess of £2.5 million;a

• there is thus some evidence that public sector executive pay is often lower than that in the private sector. But it is difficult for the public to be sure that senior pay matches performance, since there is a ‘patchwork quilt’ of pay arrangements, and good governance and transparency are not always guaranteed;

• public sector pay needs to be more focused on managing and rewarding performance. Good governance arrangements can help to ensure that public sector pay is fair – but they are unlikely to be enough on their own. There is a strong case for a maximum pay ratio. There should be disclosure requirements based on such a pay multiple and a greater emphasis on the competitiveness of executive labour markets; and

• public sector bodies that are furthest from government tend to provide the highest lead salaries. As the public sector is reformed, there is a risk that unfair pay increases may rise. Restricted labour markets for senior positions unnecessarily fuel pay inflation.

The final report will be published in March 2011, and will set out recommendations on:

• governance and transparency;

• the pay multiple;

• performance pay; and

• market competitiveness.

a) Hutton Review of Fair Pay (2010), Interim Report, p.51.

The principles of reform revisited

1.21 The Commission’s four main principles for public service pensions were chosen to provide a balanced and comprehensive framework with which to consider the case for reform in the interim report. The Commission has assessed the options for long-term structural

31


reform in the final report against these same principles. The principles reflect the range of issues covered in the evidence submitted to the Commission in its calls for evidence and stakeholder events.

Affordable and sustainable

1.22 Public service pension expenditure must be affordable. To be sustainable, it must remain affordable over time. The affordable level of pension cost is a decision for the Government within the context of a wide range of priorities. But it cannot be assessed in the short term alone since the effects of pension decisions build up and persist over decades.

1.23 Affordability should also be analysed in the context of the Government’s overall expenditure. A reform that reduces the take-up of means-tested benefits such as Pension Credit, for instance by reducing the number of people opting out from public service pension schemes, might have a substantial positive impact on long-term finances. This could not be calculated by looking at public service pension spending on its own.

1.24 In order to be sustainable, a scheme must be able to manage and share risks effectively, without dramatic increases in costs. The Commission has assessed options for change according to their ability to deliver an affordable and sustainable system, one which has the robustness and flexibility to withstand future uncertainties.

Adequate and fair

1.25 Public service pensions should provide an adequate level of retirement income for public service workers with a reasonable degree of certainty, while maintaining high levels of employee participation.

1.26 Chapter 2 sets out the Commission’s view on adequacy, informed by stakeholder engagement and responses to the interim report. Long-term structural reform options have been analysed against their ability to deliver an adequate income for different groups.

1.27 As well as providing adequate levels of income in retirement, public service pensions should be fair. This includes fairness in the distribution of contributions and benefits between members of the same pension scheme – by income, by career path and by time of entry; fairness between different schemes; fairness between generations of taxpayers; and fairness between taxpayers and the public service employees. The fairness of different scheme designs, in their distribution of both costs and risks, is discussed further in Chapter 3.

Support productivity

1.28 To support productivity, public service pension scheme design should support an efficient labour market for employees so that the taxpayer can be confident that public

32


services provide value for money. While not its primary purpose, scheme design should look to aid the recruitment and retention of the right people in the right jobs in a cost-effective and flexible way, and should usually avoid barriers to the movement of employees between sectors.

1.29 The interim report showed that existing public service pensions can be a barrier to the outsourcing and mutualisation of public services that could drive greater productivity and efficiency in public services. The current landscape on access to schemes is assessed in Chapter 5, in the light of HM Treasury’s ongoing review of the Fair Deal policy.

Transparent and simple

1.30 Public service pensions should be widely understood. Scheme members need to know their entitlements and potential future benefits. The population as a whole needs to be confident that public service pensions offer value for money. Key design features and costs to employers and employees should thus be set out clearly and transparently. It follows that public service pension design should be as easy to understand as possible. Only with good general understanding can the Commission hope to achieve the wide agreement that is central to reform.

1.31 In line with this principle, the Commission thinks it is very important that public service pension schemes, like those in the private sector, have a clear legal framework and a governance structure that outlines clearly the roles of interested parties. Chapter 6 discusses how to achieve these goals in more detail.

Application of the principles

1.32 These principles form the basis for assessing the long-term reform choices. As discussed in the interim report, the principles were not weighted and nor was the evidence associated with them. Some principles are in conflict. For example, some of the options to ensure long-term sustainability (discussed in Chapter 4) are relatively complex and therefore difficult to understand, contrary to the principle of transparency and simplicity. This report attempts to make the trade-offs plain so that the advantages and disadvantages of different options are clear.

The purpose of pensions

1.33 In light of these trade-offs, it is important to consider the purpose of public service pensions. There are many possible reasons for public service employers to provide pension benefits to their employees.10 Those that stand out are centred on adequacy and fairness and on supporting productivity.

10 Barr and Diamond (2008), Reforming Pensions, Chapter 2, discuss the purposes of pensions in general.

33


Adequacy and fairness

1.34 At the core of occupational pension schemes is the aim of ensuring that income is adequate across an employee’s lifetime. Since an extra pound is of more value to someone when he is poor than when he is rich, an individual’s lifetime welfare will usually be higher if consumption is smoothed over the life cycle, consuming a roughly equal amount each year.11 Given that people’s earning capacities are typically lower in later life, this can be a strong motivation for providing pensions.

1.35 It could be argued that individuals should just be left to make their own arrangements to smooth income in retirement. But there is good evidence that people might then make decisions which are harmful either to themselves, with undersaving resulting in inadequate resources in later life,12 or to wider society, for instance because of a reliance on means-tested benefits.13 There are thus good reasons for governments to promote consumption smoothing through pension schemes.

1.36 Insurance is a further central motivation for pensions. By providing constant pension payments throughout an employee’s lifetime, government can help to insure against the risk of low incomes for those who live for longer than they had expected. Public service pension schemes also provide an element of family insurance through survivors’ pensions and death in service benefits (partially insuring family members against the risk of the death of the pension holder).14

1.37 Governments could also have an interest in promoting poverty relief and redistribution towards the worse off through public service pensions, though the state pension system is of most importance in this respect. The Commission’s analysis suggests that, under current scheme designs, about 1 in 8 of public sector pensioners with at least 20 years’ service could expect to be eligible for means-tested Pension Credit payments at some stage in their lives.15 In these circumstances, the true fiscal cost of providing pensions to the low paid might be relatively low, because the occupational pension in part replaces other income from the state.

1.38 However, there are other ways of achieving poverty relief and redistribution within the overall pension system, particularly the basic State Pension and Pension Credit, so these are not central motivations for occupational pensions. To encourage pension saving and to avoid

11 In technical terms, individuals display declining marginal utility of consumption; see Modigliani and Brumberg (1954),

‘Utility Analysis and the Consumption Function: An Interpretation of Cross-Section Data.’ There are many potential

complications to this simple story. For instance, because housing, taxation and travel costs are typically lower in retirement

than during the working life, the consumption level required to maintain a given standard of living is usually somewhat

lower in retirement. But the broad life-cycle hypothesis has a strong theoretical basis and good empirical support.

12 Laibson (1996), ‘Hyperbolic Discount Functions, Undersaving, and Savings Policy.’

13 See Feldstein and Liebman (2002), ‘Social Security’, who describe the US social security system as a way ‘to prevent

free-riding in the presence of altruism.’

14 These ancillary benefits are discussed further in Chapter 3.

15 Based on the Pensim2 database provided by the Department for Work and Pensions (see Annex C). Most of those who

are eligible for Pension Credit payments have relatively short careers in the public sector; about 1 in 3 of all employees

with some service in the public sector could expect to receive Pension Credit payments.

34


reliance on state benefit it is important to maximise participation in public service pension schemes where this is appropriate, as discussed in Chapter 3.

Supporting productivity

1.39 Pension schemes could help public service employers to recruit and retain effective employees. In principle, providing remuneration through pensions could be more cost-effective than doing so through pay, for the following reasons:

• Selection. Because pension schemes do not provide remuneration until late in life, they might attract particular kinds of employee – those who value the future more highly, work more effectively, and are less likely to leave their jobs.16 This could make it relatively cheaper to provide remuneration though pensions instead of pay. This reason could be most important for society as a whole where the employer provides substantial training, as in the NHS and armed forces – by hiring people who are likely to stay for several years, the employer will find it easier to recoup its investment.

• Motivation. Pensions could play a role in encouraging employees to work hard. For instance, the presence of an adequate and secure pensions guarantee could support employee engagement and commitment to their employers.17

• Risk transfer. If employees value the risk transfer implicit in many pension schemes, they might be prepared to receive lower total remuneration in return. For instance, employees might prefer not to bear the risk of variation in the price of annuities at retirement (and so variation in retirement income). Most defined benefit schemes transfer this risk to the employer.

1.40 These are potentially important impacts. However, a downside of providing remuneration through pensions is that pensions lack the flexibility and responsiveness to circumstances that pay can provide. For instance, an employer who finds that a particular set of skills is in high demand can increase pay for people with those skills, or offer retention bonuses to current employees. It is much harder to target pensions in a similar fashion.

1.41 Moreover, there is evidence that the valuation of pension entitlements by employees is lower than their cost to public service employers.18 Greater transparency, discussed in Chapter 6, could help to reduce this gap, but it suggests that pensions will often not be as cost-effective in recruiting and retaining employees as other aspects of remuneration. Pension design should also guard against perverse effects. The interim report discussed the

16 Ippolito (2002), ‘Stayers as ‘Workers’ and ‘Savers’: Toward Reconciling the Pension-Quit Literature.’

17 Evidence submitted to the Commission by Towers Watson suggests motivation as a possible important reason for

public service pension provision.

18 See IPSPC (2010), Interim Report, Chapter 6.

35


effect of final salary pension schemes in reducing labour market mobility, which could harm productivity in the wider economy.

1.42 Overall, the Commission feels that public service employers have historically used pensions excessively for workforce management, a task for which they are not very well suited and in which they are unlikely to provide value for money. Public service pensions should continue to be a valuable part of the reward package, helping to attract and motivate high-quality employees.

Implications

1.43 The purpose of public service pensions has rarely been clearly articulated, perhaps because of the disparate histories of public service pension schemes and their varying roles. In its recent report on the 2007-08 reforms, the National Audit Office commented that the Treasury ‘did not make a clear statement about the purpose of public service pensions and the types of employee behaviour it wished to encourage and support through them...’19 It recommended that, in the light of the IPSPC’s recommendations, ‘the Treasury, Government departments and public service employers should agree and communicate a clear view of the purpose of public service pensions...’20

Recommendation: The Government should make clear its assessment of the role of public service pension schemes. Based on its framework of principles, the Commission believes that the primary purpose is to ensure adequate levels of retirement income for public service pensioners (Recommendation 1).

1.44 The Commission does not believe that public service pensions should be primarily concerned with workforce management. But employer contributions to current public service pensions can be worth a fifth or more of the pay bill, and pensions will continue to be an important part of remuneration. As such, public service employers and pay review bodies should consider the impact of public service pensions on recruitment, retention and workforce management. The interim report found no evidence that pay is lower for public service workers to reflect higher levels of pension provision.21 Therefore current pension schemes do not appear to offer best value for money. However, there is evidence to suggest that some highly-skilled workers earn more in the private sector, particularly in the South East and London,22 and the careful design of remuneration packages for these groups could be especially important.

19 National Audit Office (2010), The impact of the 2007-08 changes to public service pensions, p.34.

20 National Audit Office (2010), The impact of the 2007-08 changes to public service pensions, p.10.

21 IPSPC (2010), Interim Report, Chapter 6.

22 See Pensions Policy Institute (2008), An assessment of the Government’s reforms to public sector pensions, p.48 and

Institute for Fiscal Studies (2011), Green Budget 2011, Chapter 7.

36


Recommendation: Pensions will continue to be an important element of remuneration. The Commission recommends that public service employers take greater account of public service pensions when constructing remuneration packages and designing workforce strategies. The Government should make clear in its remits for pay review bodies that they should consider how public service pensions affect total reward when making pay recommendations (Recommendation 2).

37


2 The deal

Box 2.A: Summary

• The package of reforms recommended by the Commission is a balanced deal that delivers fair outcomes for public service workers and for taxpayers and builds trust and confidence in the system.

• The Government should ensure that public service pensions, along with a full state pension, deliver at least adequate levels of income (as defined by the Turner Commission benchmark replacement rates) for scheme members who work full careers in public service. Employers should seek to maximise participation in the schemes where this is appropriate. Adequate incomes and good participation rates are particularly important below median income levels.

• Government, as a large employer, is capable of bearing the majority of the risk associated with pension saving and should continue to do so. Government should continue to provide a form of defined benefit pension as the core scheme design.

• However, there should be a fairer sharing of risk between government (and ultimately taxpayers) and scheme members. It is fair for scheme members who are still working to bear more of the risks associated with changes in longevity, for example, by working longer if life expectancy increases.

• As soon as practical, members of the current defined benefit public service pension schemes should be moved to the new schemes for future service. Most schemes should link a member’s Normal Pension Age to their State Pension Age in the future. But current workers must be protected from a sudden change in their pension benefits or pension age. The Government must also honour in full the pension promises that have been accrued by scheme members.

• The taxpayer should have certainty that public service pension scheme costs are under control. The Government should set out a fixed cost ceiling for each scheme which is the proportion of pensionable pay that public service employers will contribute, on average, to employees’ pensions over the long term.

• Every public service pension scheme (and individual Local Government Pension Scheme Fund) should have a properly constituted, trained and competent Pension Board, with member nominees, responsible for the delivery of good governance and administration.

• It is important that scheme members and taxpayers understand how public service pension schemes work and how much they cost. The Government should ensure that there is transparent publication and robust scrutiny of data associated with the schemes and that this is overseen by appropriate organisations.

38


2.1 The diagram in the Foreword outlines the deal that the Commission believes will lead to a robust system for public service pensions in the future. This chapter builds on that picture. If implemented, the deal will strike a fair balance between public service workers and taxpayers in respect of public service pensions.

An adequate income

2.2 Ensuring that public service pensions continue to provide at least an adequate level of income in retirement is, in the view of this Commission, a crucial result of any structural reforms of the schemes.

2.3 In the interim report, the Commission began exploring the issue of adequacy in the public service pension context. It showed that under the current final salary arrangements, even before taking account of the lump sum payments, current public service pension schemes in conjunction with a full state pension would deliver an adequate level of income in retirement for those who work full careers in public service, based on the replacement rates1 defined by Lord Turner in his Pensions Commission report.2

2.4 In the Commission’s final call for evidence, respondents were asked how the Commission should think about the issue of adequacy and whether a full state and public service pension should ensure people reach an adequate level of income. There was a fairly broad consensus that the benchmark replacement rates set out by the Turner Commission in 2004 were the appropriate way of thinking about adequacy. Some respondents thought that these levels should be considered the minimum required. The Commission agrees with this view. The Turner Commission’s second report3 presented research evidence which suggested that the benchmark replacement rates were closer to what people thought was a ‘minimum’ income in retirement rather than a ‘comfortable’ level of income.

2.5 Similarly, most respondents felt that a full public service pension, in conjunction with a full state pension, should deliver at least an income at these levels. Again, the Commission agrees with these views and believes that the Government should ensure future public service pension schemes (in conjunction with a full state pension) deliver at least the minimum income level in retirement recommended by the Turner Commission for scheme members who have full careers in public service, and proportionately for those with part careers. This is particularly important for low earners, who are less likely to have any other assets and income sources that they can rely on in retirement. This will mean that those members who work a full career in public service and retire on below median incomes should receive at least two thirds of their final salary from a public service pension and full state pension.

1 These replacement rates measure the level of income immediately after retirement compared to the level of income

immediately before retirement.

2 Pensions: Challenges and Choices The first report of the Pensions Commission, 2004.

3 A New Pension Settlement for the Twenty-First Century: The Second Report of the Pensions Commission, Appendix D,

2005.

39


Chart 2.A: Gross replacement rates delivered by typical current public service pension schemes and state pension system

0%

40%

80%

120%

160%

Per

cen

t

Pre-retirement income (£)

Gross replacement rate - full career

Gross replacement rate - excluding lump sum

Benchmark replacement rates

Source: IPSPC analysis.

2.6 Chart 2.A shows that current schemes deliver these adequacy levels4 when combined with a full state pension after a lump sum has been taken. These replacement rates should be regarded as a minimum. To ensure that a comfortable level of income can be attained under the current schemes, scheme members who have worked full careers in public service would need to consider taking a smaller lump sum or have other income sources in retirement.

2.7 Another key factor when thinking about adequacy is the level of participation in the schemes. There is little point designing a pension scheme that delivers adequate income levels, if public service employees, for one reason or another, decide not to remain members of the scheme once they are enrolled into it or decide not to join at all. At the moment around 85 per cent of public sector workers are members of an employer sponsored pension scheme. But this varies by income level. Chart 3.H shows that as earnings increase so do participation rates in public sector schemes. It is important that the future structure of public service pension schemes maintains or improves the participation rates of employees, especially below median earner income levels. Important considerations that will affect participation rates in the future are the level of employee contributions required, the trust that scheme members have in the scheme and the ease with which members are able to understand the scheme.

4 This is the level of income from a typical final salary public service pension scheme – a 1/60th accrual rate will deliver a

pension of two thirds of final salary before any lump sum is taken and half of final salary after the maximum lump sum

has been taken for a member who has 40 years of service.

40


Box 2.B: Automatic enrolment from 2012

The previous Government introduced plans for all eligible workers not currently in a qualifying workplace pension scheme to be automatically enrolled into one, as a way of supporting saving for retirement.

Automatic enrolment will start to be implemented from October 2012, although a reviewa carried out for the current Government recommended that large employers who wish to enrol their eligible employees automatically ahead of the start date should be able to do so, from July 2012.

A feature of these reforms is that employees who have opted out of schemes will be automatically re-enrolled into schemes around every three years, which should help to increase participation in pension schemes, including public service schemes, in the medium term.

a) The Pensions Bill 2011 entered the House of Lords in January 2011 and includes measures to implement the recommendations of the Making Automatic Enrolment Work Review

Recommendation: The Government should ensure that public service pension schemes, along with a full state pension, deliver at least adequate levels of income (as defined by the Turner Commission benchmark replacement rates) for scheme members who work full careers in public service. Employers should seek to maximise participation in the schemes where this is appropriate. Adequate incomes and good participation rates are particularly important below median income levels (Recommendation 3).

2.8 The level of income delivered by and participation in public service pension schemes are not the only important aspects to consider when looking at adequacy: certainty of income is also important.

2.9 As such, current workers must be protected from a sudden change in their pension benefits or pension age, particularly when they are close to retiring. The issue of transition is considered in more detail in Chapter 7.

2.10 Certainty of income will also be important in the future, especially for people with lower incomes. Therefore, it is crucial that risks associated with pensions are carefully considered when looking at potential new scheme designs.

Defining the risks

2.11 There are many different types of risk involved in pensions. Occupational pension schemes can be designed in ways that share risks differently between scheme members and the sponsoring employer, which is usually the Government and ultimately the taxpayer in the public service pension schemes. By looking at the different types of risk and making a judgement, informed by the principles outlined in Chapter 1, about who is best placed to

41


bear these risks, the Commission has been able to develop a shortlist of potential options for structural reform.

Identifying the risks in public service pension schemes

2.12 The types of risk in pension schemes can be categorised in many ways and vary by scheme structure, funding status and by the type of member. In Table 2.A below, the Commission’s interpretation of the key risks in the context of a defined benefit (DB) scheme are outlined. The risks are presented from the point of view of the employer, who bears the majority of risks in this type of scheme. They are also presented as downside risks – so risks that can increase the costs of running the schemes – although costs could fall if the risks move in the opposite direction to that presented here.

Table 2.A: Risks in defined benefit pension schemes

Type of risk Description of risk

Investment For funded schemes, this is the risk that lower than expected asset returns mean that there are insufficient assets to pay benefits when they come into payment. For unfunded schemes, the risk is that GDP (and therefore tax revenues which fund future pension commitments) does not rise as quickly as projected.

Inflation Risk that higher than expected price inflation increases the cost of providing pensions.

Salary Risk that higher than expected salary increases increase the cost of providing pensions.

Longevity Risk that higher than projected longevity increases the cost of providing a defined level of benefit. Longevity changes can happen before or after retirement.


2.13 Defined contribution (DC) schemes, such as the Civil Service Partnership scheme, which are more typically found in the private sector are subject to a different set of risks. These are set out below, largely from the point of view of the members who typically bear most of the risks. The main exception is post retirement longevity risk which sits with whoever is paying the annuity, which is often an insurance company.

42


Table 2.B: Risks in defined contribution pension schemes

Type of risk Description of risk

Investment The risk that lower than expected returns on contributions paid provides a lower than expected pension fund from which to buy an annuity at retirement.

Interest rate/annuity risk The risk that adverse market movements close to retirement reduces the amount of pension that can be bought from a fixed sum of money.

Inflation The risk that higher than expected inflation erodes purchasing power of benefits after they come into payment.

Longevity Before retirement: Risk that higher than expected life expectancy reduces the amount of pension that can be paid from a set amount of assets.

After retirement: Risk to the employer / insurer that higher than expected life expectancy increases the cost of providing a defined level of benefit.


2.14 As indicated from the descriptions above, some of these risks apply before retirement only (such as salary risk in a DB scheme), some apply in retirement only (inflation risk in a DC scheme) and some risks can occur before or after retirement (such as changes in longevity).

2.15 These risks have been identified as technical risks that impact on scheme design. The rest of this section focuses on these risks. Of course, pension schemes are inherently subject to many others that require consideration. Among them are governance and regulatory risk, the impact of demographic and behavioural changes and default risk (the risk that for one reason or another, possibly employer insolvency, the full pension is not paid out).

Current public service pension schemes: who bears the risk?

2.16 Within pension schemes the three main parties that may bear risk are the employer, the scheme members and in the event of insuring pension incomes, the insurer. Across the public services, almost all pension schemes are run on a defined benefit basis, and most of those provide benefits based on final salary. This places the majority of the risks in the hands of the employers – or in this case, mainly the Government and ultimately the taxpayer.

2.17 After retirement, DB schemes provide members with a pre-defined amount which is paid throughout the life of the member (and a surviving spouse) and indexed throughout this period in line with inflation. How much this benefit ends up costing the employer depends on both financial and longevity factors and the risks relating to these are borne by the employer.

43


2.18 Before retirement, the risks are centred on how pension builds up and how much employees and employers respectively pay towards it. In a final salary scheme, the employer bears some salary risk, since those who retire on a higher than anticipated salary end up costing the employer more than they had provisioned for. In final salary schemes, members will have paid contributions to the scheme based on something similar to their average salary, but will receive benefits based on their final salary. Taxpayers will pick up the cost of the difference between average and final salaries and members will benefit where final salaries are higher than average salaries. This effect will be particularly visible where people have experienced rapid salary growth. In average salary schemes, members bear more of the risk – salary levels throughout a member’s career will determine their income at retirement as well as their contributions to the scheme.

2.19 The employer can also manage costs through the contributions it requires of its active members. In some current public service pension schemes ‘cap and share’ mechanisms have been introduced to share longevity risk with scheme members. This allows increases in costs caused by improvements in longevity to be shared by employees and employers up to a fixed employer cap. Above the cap increases are passed to employees.

2.20 In DC schemes, most pre-retirement risks fall on the member. Members are subject to the risk that investments in their fund do not perform well. At retirement the amount of pension the fund is able to finance is subject to longevity risk and interest rate risk. However, once an income stream is secured the investment and longevity risk passes to the insurer (if bought out) or the employer (if retained inhouse). If a member opts for a flat rate annuity which does not increase in line with changes in prices, they will also bear inflation risk in retirement.

Future public service pension schemes: who should bear the risk?

2.21 The question of who should bear the risk in pension schemes is subjective and complex. The Commission’s analysis, based on balancing the key principles, has led to the following conclusions.

2.22 It is not unreasonable to treat different groups differently, given the variations in capacity to bear risk. Younger members and those with higher earnings are arguably more able to take on risk given the ability to respond to adverse outcomes. Younger members have more time to make good any shortfalls in pension saving caused by changes in longevity or investment performance. Higher earners are likely to have other assets or income streams which they can rely on in retirement. However, the ability to bear risk must be differentiated from the ability to suffer losses. Younger members may have more capacity to bear risks, but that does not mean they should expect to be worse off as a result.

2.23 An extension of the age distinction is that retired members have little capacity to bear and manage risks, and as such, it is right for most risk after a member has retired to be with the employer (or insurer).

44


2.24 Salary risk and longevity risk before retirement are areas where increases in the cost of pension provision benefit members at an individual level, whether that is through higher pension income or pension income paid for a longer period. As such it is the Commission’s conclusion that it is reasonable for employers to share these risks with the members.

2.25 Conversely, increases in pension costs because of uncertain asset returns or high interest rates or inflation are generally areas where members individually have little control and do not benefit. Large employers, such as government, have a better ability to bear these risks. For this reason, the Commission believes that these risks should not be passed to the member.

Table 2.C: Summary of the Commission’s view of who can bear risks in public service pension schemes

Younger members Older members Pensioner members

Longevity risk Longevity risk Longevity risk

Lower earners Investment risk Investment risk Investment risk

Salary risk Salary risk

Longevity risk Longevity risk Longevity risk

Higher earners Investment risk Investment risk Investment risk

Salary risk Salary risk

Key: Cannot bear risk Might be able to bear risk

Can bear risk


The implications for scheme design

2.26 The Commission has set out what it considers the optimal balance of risks should be against where risks sit in current public service pension schemes and under several alternative scheme structures. More information on this is included in Chapter 3.

2.27 The results show that final salary structures are not a good fit, primarily because of the salary effects. Defined contribution schemes are also a poor match as an occupational pension scheme offered by the Government as a large employer, leaving too much pre-retirement risk in the hands of the members which can be borne more efficiently by the employer.

2.28 The analysis shows that the most appropriate structures, at least for the majority of scheme members, have a defined benefit design.

45


Recommendation: The Government should continue to provide a form of defined benefit pension as the core scheme design (Recommendation 5b).

2.29 However, the Commission’s analysis shows that at present under the existing final salary schemes too many risks are faced by government and the taxpayer. As such, it is the Commission’s view (outlined in more detail in Chapter 7) that all members of the current defined benefit public service pension schemes should be moved to the new schemes for future service.

Accrued rights

2.30 In the interim report, the Commission showed the rights which have already been accrued by public service pension scheme members will make up the majority of benefits being paid for the next quarter of a century. This analysis is reproduced in Chart 2.B below.

Chart 2.B: Gross benefit expenditure by type of memberUNCLASSIFIED

0.0%

0.2%

0.4%

0.6%

0.8%

1.0%

1.2%

1.4%

1.6%

1.8%

2.0%

2009/10 2019/20 2029/30 2039/40 2049/50 2059/60

Perc

enta

ge o

f pro

ject

ed G

DP

Year

Existing pensioners and deferreds Current employees - past service

Current employees - future service Future employees

Source: IPSPC interim report, Chart C.8.

2.31 Despite the fact that it may constrain options for reform, it must be a fundamental principle of any framework for public service pension schemes that the accrued rights of existing scheme members are not reduced. This raises the issue as to what constitutes an accrued right in the context of the public service pension schemes, and therefore which rights should be protected when implementing any change to the future benefit structure. More information on this issue is in Chapter 7.

46


2.32 The Commission is clear that pensions in payment will not be affected by this review. In addition, current active members of final salary public service pension schemes should have their pension from the current schemes based on their final salary, rather than on their salary at the point they move into the new schemes for future accrual. This provides a level of transitional protection for scheme members, which should help to build trust and confidence in the process of reform, while allowing the introduction of new pension arrangements for the future which are seen as fair to members and taxpayers.

Managing the cost to taxpayers

2.33 Guaranteeing that public service pension schemes are designed to deliver at least adequate levels of income in retirement will help maintain the trust that scheme members have in the schemes and minimise the need for means-tested benefits for public service pensioners. However, taxpayers also need to be reassured about the future sustainability of the schemes.

2.34 Linking a member’s Normal Pension Age to their State Pension Age in most schemes (discussed in more detail in Chapter 4) will share the impact of rising longevity in the past and expected changes in the future more fairly. Ensuring that employer contributions to the public service pension schemes remain below a fixed cost ceiling will result in a degree of certainty about the future nature of the pensions promise for taxpayers and act as an extra safety valve in the system. The Government, on behalf of the taxpayer, should set out a fixed cost ceiling - the proportion of pensionable pay that they will contribute, on average, to employees’ pensions over the long term.

2.35 To some extent this approach has already been encompassed within current scheme designs with the cap and share mechanisms. Although these mechanisms have not yet been fully tested in practice, the NHS scheme operated an employer cost cap when they undertook the 2008 reforms. A cost ceiling would effectively replace the current cap and share mechanisms. An employer cost ceiling could factor in automatic adjustments to maintain a fixed cost level.

2.36 There are a number of different ways that a cost ceiling could be defined and there will be a decision for the Government to make about the level of the ceiling. There are also different ways of adjusting different types of schemes so that the ceiling is maintained. These issues are discussed in more detail in Chapter 4.

Scheme governance and transparency

2.37 An additional mechanism to help aid long-term sustainability is for the appropriate information about future costs to be readily available. Currently there is inconsistency in what scheme data and valuations are published and such information is often difficult to access. A minimum standard should be determined, to include the appropriate financial data, but also membership and valuation details across the schemes. It is the Commission’s view,

47


outlined further in Chapter 6, that the Office for Budget Responsibility should consider the sustainability of public service pensions when considering long-term fiscal sustainability and should ensure there is appropriate information available for them to make that assessment.

2.38 The current arrangements for the governance of public service pension schemes require improvement to ensure that there is a fair balance between the interests of scheme members and taxpayers. Communication with scheme members should also be improved and the future schemes should be simple enough for members to understand.

48


49


3 The design

Box 3.A: Summary

• The Government should adopt a scheme based on career average earnings as the core public service pension scheme, to ensure there is an effective balance of risks between scheme members and employers. A single benefit design should apply across the whole income range.

• The Commission is not recommending specific accrual rates, indexation levels or employee contributions. These are matters for the Government as they impact on affordability. Accrual rates, indexation methods and employee contribution levels do need to be considered together to ensure that pension schemes deliver at least adequate incomes, maintain high participation rates and are sustainable.

• Benefits should be indexed by earnings before retirement for active members of future public service pension schemes, to support adequacy of pension income at the point of retirement. The choice of either prices or earnings indexation for deferred members should depend on the Government’s assessment of the role of public service pensions and the balance between the competing principles of adequacy and fairness and supporting productivity. Post-retirement, pensions in payment should be indexed in line with prices to maintain their purchasing power and adequacy during retirement.

• Caps on indexation, either pre or post retirement, should not be introduced.

• When setting member contribution levels the trade off between affordability and the impact of opt outs on adequacy should be considered by the Government. The differing characteristics of higher and lower earners, including that higher earners tend to live longer, should be addressed through tiered contribution rates.

• Members should be provided with greater control over when to start drawing their pension benefits. Lump sums should continue to be made available to members of public service pension schemes through commutation.

• Actuarial reduction and enhancement should be applied to early and late retirement, and caps on pension accrual should be removed or significantly lifted. Flexible retirement should be encouraged and abatement of pensions in its current form for those who return to work after drawing their pensions should be eliminated. The costs of these changes will need to be assessed and considered as part of the reform package.

• Members should be encouraged to make additional contributions to their pensions in a simple and transparent manner.

• Ancillary benefits should be broadly retained in their present form. The current system provides people with equal access to a range of valued, cost-effective benefits on top of their pension, which provides a useful safety net.

50


From long-list to short-list of scheme design options

3.1 There are a huge number of potential different pension scheme designs, ranging from final salary defined benefit (DB) schemes to pure defined contribution (DC) schemes. In between there is a myriad of different approaches and hybrid options. The Commission identified seven distinct types of scheme (excluding hybrid schemes) to consider, which are shown in Table 3.A.

Table 3.A: Scheme designs

Type of scheme Description

Final salary The pension at retirement is defined, based on the number of years of service and the member’s final salary at retirement. Once in payment, the amount of pension is guaranteed.

Career Average Revalued Earnings (CARE)

The pension at retirement is defined, based on the number of years of service and the member’s average revalued salary. Once in payment, the amount of pension is guaranteed.

Cash balance with guaranteed conversion terms

The pension pot at retirement is defined, based on the proportion of salary set aside each year and the guaranteed rate of interest earned. The pot is converted to pension on guaranteed terms that are set by the scheme at an agreed point before retirement. Once in payment, the amount of pension is guaranteed.

Cash balance with open market annuity

The pension pot at retirement is defined, based on the proportion of salary set aside each year and the guaranteed rate of interest earned. This is converted to pension on open market terms (open market annuity rates). Once in payment, the amount of pension is guaranteed.

Notional defined contribution

The pension pot at retirement depends on the amount of contributions paid in and the return on the notional assets (no investments are actually made – the performance of a notional portfolio is tracked). This is converted to pension on open market terms (open market annuity rates). Once in payment, the amount of pension is guaranteed.

Collective defined contribution

The pension pot at retirement depends on the amount of contributions paid in and the return on the invested assets. Returns are smoothed since risk is shared amongst all participants in the scheme. The pension pot is converted to pension on open market terms (open market annuity rates). Risk sharing continues post-retirement, with conditional indexation of pensions in payment dependent on the financial health of the fund. The pension in payment can therefore increase or decrease.

Individual defined contribution

The pension pot at retirement depends on the amount of contributions paid in and the return on the invested assets. This is converted to pension on open market terms (open market annuity rates). Once in payment, the amount of pension is guaranteed.


51


3.2 Different types of schemes can be designed so that they give the same, or similar, outcomes for scheme members. As an example, a cash balance scheme with a guaranteed conversion factor can be designed to replicate a CARE scheme by choosing an appropriate accrual rate and pension conversion factor. On the other hand, many of these scheme types can be designed in ways that would lead to different outcomes for members. For example, a DC scheme could have returns linked to the market or guaranteed investment returns underwritten by the employer.

3.3 The choice of scheme, and the choice of parameters within a scheme, therefore has a large impact on how the scheme looks and feels to the member, and how risks are shared between members and government. Since outcomes are more important to the Commission than the type of scheme, in the first instance the scheme types were rated against how well they matched up with the Commission’s view of who should bear the key risks, as set out in Chapter 2 (pre-retirement longevity risk is considered in Chapter 4):

• Salary risk: The risk of higher than expected salary increases should lie with the member.

• Investment risk: Government, as a large employer, is better placed to bear investment risk than members.

• Post-retirement longevity risk: Pensioners have little scope to manage variation in their incomes post-retirement, and so government should bear this risk.

3.4 Table 3.B shows the results of the Commission’s risk analysis, which was used to generate a short-list of possible scheme designs for further consideration.

52


Table 3.B: Risk analysis of different scheme designs

Type of scheme Who bears the salary risk?

Who bears the investment

risk?

Who bears the post-retirement longevity risk?

Conclusion

Final salary Government Government Government Rejected

CARE Member Government Government Short-listed


Member Government Government Short-listed

Cash balance with open market annuity

Member Government Government Short-listed

Individual defined contribution

Member Member Government Rejected

Notional defined contribution

Member Both Government Rejected

Collective defined contribution

Member Member Member Rejected

Key: Risk borne by wrong party Risk borne by correct party


3.5 Schemes where the amount of pension is based on a member’s final salary were ruled out because of the finding in the interim report that they were unfair between scheme members. Those members who experience more rapid wage growth benefit disproportionately from these types of schemes, as set out in Chapter 1. Salary risk is retained by government, whilst the Commission believes this type of risk should be shared with scheme members.

3.6 The Commission also took the view that members are not better placed to bear investment risk than government as a large employer, especially at lower incomes where the uncertainty of pension amount is challenging to manage. Additionally, a funded DC scheme would create cash flow implications for government, while a notional DC scheme with returns linked to investment market performance would introduce extra investment risk into the system for government and scheme members. Funded or notional DC schemes have therefore been ruled out as the core scheme.

3.7 Collective defined contribution schemes, where the amount of pension payable can be changed after retirement in response to longevity or investment performance, were rejected because the Commission took the view that there should be certainty about the level of income post-retirement. Pensioners (particularly older pensioners) have little scope to supplement their income once they have retired.

3.8 This left a short-list of scheme designs where the pension is related to average earnings over the whole of a member’s public service career (rather than final salary), where the

53


scheme has defined ‘investment’ returns and where there is certainty for the member in retirement.

3.9 Career average schemes meet these criteria. Salary risk is borne by members and the defined benefit arrangement provides the recommended risk structure after retirement. There is no in-built way of managing longevity risk pre-retirement in career average schemes but the general career average structure could be combined with a risk-sharing mechanism to manage this risk (these mechanisms are considered in Chapter 4).

Box 3.B: How a CARE scheme could work

In a Career Average Revalued Earnings scheme, each year the scheme member earns an amount of pension based on the scheme accrual rate and their salary in that year. This is then revalued (or indexed) each year. At retirement each year’s accrual is summed up to give the total pension. The parameters used here are purely illustrative and are not recommendations by the Commission.

Consider Alice, who has 3 years of service in a CARE scheme with a 60ths accrual rate and 3 per cent revaluation. On joining the scheme she had a salary of £30,000 and this grew with pay rises each year.

Care example

Alice retires after 3 years with a pension of £1,605 a year.

Under a final salary scheme with the same 60ths accrual rate, Alice would have ended up with a similar level of pension because the increase in her salary (4 per cent per year) is around the same as the level of indexation in the CARE scheme.

3.10 Cash balance schemes go one step further. As well as having the advantages of career average mentioned above, there is the potential to share longevity risk pre-retirement with members in a more straightforward manner than in career average schemes.

3.11 Under cash balance schemes it is the size of the pension pot at retirement that is defined, not the amount of annual pension. The pension pot is then converted to an annual pension using a pension conversion factor (a factor which determines how much money is needed in the pot to purchase £1 of annual pension income). The conversion terms

Accrual Rate x Salary = Pension

Revalued in 2012

Revalued in 2013

Value at retirement

Year 1 (2011) £30,000 x 1/60 = £500 £500 + 3% = £515 £515 + 3% = £530 £530

Year 2 (2012) £31,200 x 1/60 = £520 n/a £520 + 3% = £535 £535

Year 3 (2013) £32,450 x 1/60 = £540 n/a n/a £540

Alice’s total annual pension at retirement £1,605


54


Box 3.C: How a cash balance scheme with guaranteed conversion terms could work

In a cash balance scheme, each year a proportion of the scheme member’s salary is paid into a pension pot. This pension pot then grows each year with guaranteed interest until retirement. At retirement the pension pot is converted into an annual pension using a pension conversion factor. The parameters used here are purely illustrative and are not recommendations by the Commission.

Consider Bob, who has 3 years of service in a cash balance scheme in which 1/3rd of his salary is paid into his pension pot. His pension pot grows with 3 per cent interest each year, and at retirement £1 of annual pension costs £20. On joining the scheme he had a salary of £30,000 and this grew with pay rises each year.

Cash balance example

Bob retires after 3 years with a pension of £1,605 a year.

could be guaranteed when the member joins the scheme, guaranteed at some fixed point before Normal Pension Age (NPA) or set using annuity rates found on the open market at the point of retirement. Under the first option, all the longevity risk in respect of current members sits with government, while under the annuity option the pre-retirement longevity risk sits with the scheme members. If the conversion terms are initially flexible and are then set at some point before NPA, then longevity risk sits with member until the terms are guaranteed and with government after that point.

Accrual Rate x Salary = Pension

Revalued in 2012 Revalued in 2013 Value at retirement

Year 1 (2011)

£30,000 x 1/3rd= £10,000 £10,000 + 3%= £10,300 £10,300 + 3%= £10,600 £10,600

Year 2 (2012)

£31,200 x 1/3rd= £10,400 n/a £10,400 + 3%= £10,700 £10,700

Year 3 (2013)

£32,450 x 1/3rd= £10,800 n/a n/a £10,800

Bob’s pension pot at retirement: £32,100

This is converted into an annual pension: £32,100 / £20 = £1,605

Bob’s total annual pension at retirement: £1,605


55


Core design

3.12 The Government has a choice in the type of scheme it can adopt as the core model for public service pensions. The Commission believes this choice should be between the types of scheme shortlisted above.

3.13 The main considerations in arriving at this final decision should be:

• how much certainty should be provided for scheme members at different points in their careers;

• the level of flexibility the Government wishes to have over pension costs; and

• how understandable different designs are for scheme members.

The balance between certainty and flexibility

3.14 The balance of risks between members and government will determine both the amount of certainty available to members, which is an important feature of the adequacy principle, and the amount of flexibility government has to control costs, which is important for the affordability and sustainability principle. Certainty is an attractive design feature for members, since it leads to transparency and ease of understanding, enabling them to plan for their retirement. It also reduces the risk to the Exchequer of inadequate provision (the danger of members needing further support from the benefits system in retirement). However, certainty for members causes issues for government around the affordability and sustainability of pension schemes in situations where the cost of providing this certainty is greater than expected. Flexibility in scheme design may therefore be an attractive feature for government. There is a key trade-off between the adequacy and the affordability and sustainability principles in this area.

3.15 The interim report identified that the main driver of increased costs for public sector pension schemes in recent years was members living longer than expected. Government had effectively provided members with certainty that their pension would be paid for life at retirement, and when this cost more than expected, government bore the bulk of the increased cost. One of the reasons government bore most of this extra cost was because, until the reforms of the last few years, the existing schemes provide little flexibility for government to pass on any additional costs to members, other than by raising contributions paid by active members (who may not be those who benefitted from the certainty provided by government).

3.16 Members’ desire for certainty and government’s willingness to provide it in public service pension schemes may depend upon the scheme members’ income levels, as well as their age. For older members with lower incomes, certainty may be the main driver, both for members when planning for retirement, and government in assuring adequate levels of income in retirement and less reliance on state benefits. Government may be less willing

56


to guarantee certainty for scheme members with higher incomes, at least on their entire pension, since these people are more likely to have other assets and income sources available to them in retirement.

3.17 Age is also an important factor. As people approach retirement, awareness of their retirement income may well increase. Although younger members may be less aware of what their retirement income may be, that is not to say they do not value certainty. It is likely however they will have a different perspective than those approaching retirement. Therefore a limited degree of certainty may be sufficient for younger members, with certainty increasing as retirement approaches so that people can plan for retirement with some peace of mind about their income level.

Understanding pension scheme designs

3.18 It is important that members are able to understand the scheme, so that they can plan for retirement. In order to gain an insight into the views of public service pension scheme members, and gauge how easy different scheme designs were to understand, the Commission invited a research team to undertake consumer research1 with public service workers. This was in the form of a deliberative workshop and there were 89 attendees with a broad diversity in terms of gender and representation across the 6 major public service pension schemes. Slightly more than half of the attendees were aged 46-59, with an under-representation of younger workers and those in part-time and manual positions. It is likely that the attendees had a higher level of pensions knowledge, understanding and expertise than might be typically expected, due to the level of trade union representation at the event and the age profile of the attendees.

3.19 Research participants were asked to consider the relative merits of presenting pension benefits in terms of either the amount of pension that had been earned (the way a CARE scheme is usually presented), or as the ‘pot’ of money available to purchase a retirement income on guaranteed terms (the way a cash balance scheme with guaranteed conversion terms is normally presented). On balance, most participants felt that the amount of pension approach was a more helpful concept than the pension ‘pot’. This was because such pots of money could not actually be accessed by scheme members, and because people’s overwhelming priority was to plan their retirement by understanding what their pension income would be, irrespective of how this is actually calculated.

3.20 When surveyed, the attendees expressed a strong preference for final salary schemes, with 83 per cent of attendees stating that this was their preferred choice. In their view, final salary schemes were trusted and felt more certain than other options. Many felt that final salary arrangements were an important part of the package they signed up to in becoming a public service worker.

1 Public service pensions reform: findings from consumer research, PwC, March 2011.

57


3.21 Once final salary and defined contribution schemes were removed from the list (since these scheme designs had been ruled out by the Commission in its interim report), 40 per cent of people stated that career average was their preferred scheme design of those available. When asked to identify the scheme that was easiest to understand from the same shorter list, 60 per cent of the attendees chose career average. More detailed results are shown in Table 3.C.

Table 3.C: Results of deliberative workshop survey

Of the following schemes, which would you say you have a preference for?

Of the schemes on this shorter list, which would you say you have a preference for?

Of the following schemes, which would you say is easiest to understand?

Final salary 83% n/a n/a

Career average 4% 40% 60%


2% 8% 4%

Cash balance with variable conversion terms

0% 0% 3%

Defined contribution 1% n/a n/a

None of the above 2% 37% 10%

I don’t understand any of them enough to make a choice

4% 10% 10%

Not answered 2% 4% 12%

Source: Survey results from IPSPC deliberative workshop.

3.22 The findings of the deliberative workshop suggest that members might find a CARE scheme preferable to a cash balance scheme, both in terms of ease of understanding and in helping them to plan their retirement. However there are a number of other factors that should be considered before making a choice between CARE and cash balance.

The choice between CARE and cash balance

3.23 CARE and cash balance schemes can be set up to provide very similar benefits. The main difference between the two is that in a cash balance scheme, while the ‘pension pot’ at retirement is defined in the rules, the amount of pension this will buy does not have to be defined. In a CARE scheme it is the pension that is defined.

58


3.24 A cash balance scheme therefore provides additional flexibility for government as they can vary the pension conversion factor (the factor which converts the cash pot to a pension amount) to reflect changes in the cost of providing pensions.

3.25 However, this flexibility will come at the cost of reduced certainty to members as they may not be able to predict their pension in retirement. This could be mitigated by providing members with regular pension projections, and providing members guaranteed conversion terms a defined period of time prior to retirement.

3.26 It is the view of the Commission that scheme members should be provided with a good level of certainty when it comes to the amount of their pension. As such, linking conversion factors to open market annuity rates, or only guaranteeing the conversion terms at some point after a member has joined the scheme, would not be appropriate solutions for public sector cash balance schemes.

3.27 In a CARE scheme there is no equivalent in-built way of managing costs, but the general career average structure could be combined with a risk-sharing mechanism (see Chapter 4) to manage this risk. It is likely that this approach will provide less flexibility for government, but would still provide an effective solution.

3.28 The Commission’s view is that the initial attraction of a cash balance approach (that it provides greater flexibility for government, which will increase affordability and sustainability) is outweighed by the need for certainty for scheme members (which will help ensure the adequacy principle is met). In addition, our research shows that a cash balance approach is harder for members to understand than a straightforward career average scheme.

Recommendation: A new career average revalued earnings (CARE) scheme should be adopted for general use in the public service pension schemes (Recommendation 7).

59


Box 3.D: The distributional impact of moving from final salary to CARE

The Commission asked the Pensions Policy Institute (PPI) to undertake modellinga regarding potential scheme designs. The aim was to determine benefit designs that are, on average, of equivalent expected average value to scheme members as the current final salary arrangements.

The chart below shows the effective employee benefit rate (EEBR), the value of the pension benefit, net of employee contributions, accrued annually by an average member of the scheme expressed in terms of a percentage of pay, for both a proxy to the current final salary schemes and an example CARE scheme.

The distributional impact of moving from final salary to CARE

0%

5%

10%

15%

20%

25%

30%

Low Flyer Mid Flyer High Flyer

Effe

ctiv

e em

ploy

ee b

enef

it ra

te

Proxy to current final salary schemes Career average scheme

Under the example CARE scheme the EEBR is higher for a low flyer than in the proxy to the current final salary schemes, while for a high flyer it is lower. A CARE scheme therefore redistributes pension benefits from high flyers to low flyers. This redistribution occurs because, for a fixed cost, the savings from the removal of the final salary link for high flyers can be recycled into providing a better level of benefit for other scheme members.

a) Further details of the modelling are given in Annex C, including definitions of low, mid and high flyers.

Higher earners and hybridity

3.29 Evidence submitted to the Commission suggests that higher earning individuals are better positioned than lower earners to bear some of the risks associated with pension provision, particularly investment risk. Due to their higher disposable incomes, higher earners have a greater capacity for self-provision of pension benefits than lower earners. They also report having a higher level of knowledge about pensions than people on lower incomes,

60


which means they may be more capable of making important decisions about their pension provision.2

3.30 It is also likely that they will have additional forms of savings and investments that they can use to provide capital and income in retirement, as Chart 3.A shows.

Chart 3.A: Distribution of household wealth excluding pension wealth in 2006-08

< £100£100 to < £200£200 to < £300£300 to < £400£400 to < £500£500 to < £600£600 to < £700£700 to < £800£800 to < £900

£900 to < £1000£1000+

Wealth, £000s

Inco

me

per

wee

k

Median Mean

Source: Wealth and Assets Survey, Office for National Statistics.

Note: The ONS advises that these figures should be treated with caution. The breakdown is only by earned income, so

sources such as state pensions are not included as income.

2 DWP (2009) Attitudes to pensions: the 2009 survey, p.46.

61


Box 3.E: Changes to pensions tax policy

High earners are likely to be affected by changes to the pension tax relief system which were announced by the Government on 14 October 2010.

The changes relate to the Annual Allowance, the amount of pension funding eligible for tax relief each year. From April 2011, this will be reduced to £50,000 from the current level of £255,000.a

The tax charge for exceeding the Annual Allowance will be a tailored charge, to cover the full marginal rate of relief from which an individual has benefited above the allowance. This is expected to follow current principles by adding the value of any excess provision to an individual’s income for tax purposes. The Government is to explore options which will allow individuals, in some circumstances, to pay a portion of any tax charge incurred for exceeding the Annual Allowance out of their pension entitlement, rather than current income.

Many individuals who are currently in final salary schemes could face a tax charge under the new regime. This is most likely to happen if scheme members are high earners or experience a significant increase in their pensionable pay in one year, perhaps as a result of a promotion. This is because the amount of pension received from the scheme depends on final salary, and so increases in pay lead to increases in the member’s pension benefits.

Moving from a final salary scheme to a career average type scheme is likely to mitigate this effect over time, particularly for those workers who have large increases in pay, since the link between pay increases and the increase in pension benefits will have been broken.

a) This amount will initially be fixed but might be indexed over the longer term. Where individuals exceed the Annual

Allowance in a given year, unused allowance from up to 3 previous years will be available to offset against the excess

provision. The fixed factor for valuing Defined Benefit accrual to test against the Annual Allowance will be 16 and the

‘opening value’ of an individual’s Defined Benefit rights will be indexed, likely to be in line with CPI.

3.31 The Commission has considered options for different models of scheme designs at different income levels, and a single scheme across all income levels, against its four principles. Options for different treatment of high earners considered by the Commission include:

• a cap on pensionable pay in the core DB scheme, with self-provision above this level;

• a cap on pensionable pay in the core DB scheme, with employer-provided funded3 defined contribution above this level;

• a cap on pensionable pay in the core DB scheme, with a less generous DB scheme above this level; and

• a single scheme across all income levels.

3 The Commission has rejected notional DC schemes with returns linked to investment market performance as it

introduces extra investment risk into the system for government.

62


3.32 The level of any cap would need to be decided. Table 3.D shows the proportion of scheme members that would be affected by the introduction of a cap on pensionable pay at various levels. A quarter of members would be affected if schemes were capped once pensionable pay was at £35,000 per year. Only 1 per cent of members would be affected by a cap set at £100,000 of pensionable pay per year.

Table 3.D: Estimated proportion of members affected by a cap on pensionable pay

Theoretical pensionable earnings cap

Proportion of scheme members affected

£30,000 36%

£35,000 23%

£40,000 13%

£45,000 8%

£50,000 5%

£75,000 2%

£100,000 1%


Note: Schemes covered are NHS (England and Wales), LGPS (all of UK), police (England and Wales) (as at 31 March

2008), teachers (all of UK), armed forces, principal civil service (UK and Northern Ireland), judiciary.

3.33 A cap on pensionable pay could have short-term implications for government cash flow as contributions would not be charged, or would need to be reduced, on any pay in excess of the cap. The Commission estimates that losing contributions above a cap set at £50,000 of pensionable pay would see a reduction of contributions into the unfunded schemes of around £2.2 billion per year.4 If the cap was set at £75,000 of pensionable pay then the contribution reduction would be in the order of £0.9 billion per year. Alternatively, if contributions continued to be charged on total pay then this would likely lead to significant opt-out among higher earners, potentially leading to an even greater reduction in the contributions paid into the unfunded schemes.

3.34 Modelling5 undertaken by the PPI on behalf of the Commission has determined that capping pensionable earnings at £75,000 has a negligible impact on the benefits that can be offered below the cap (by recycling the savings into providing better benefits on earnings under the cap). This is due to the small number of people that would be affected by a cap at this level compared to the number of people below the cap. In order to have a meaningful impact the cap would have to be set at a much lower level. This would bring into question

4 Calculations based on current salary levels and contribution rates after the introduction of the 3 per cent average

increase in employee contributions that have been announced by the Government.

5 See Annex C for further details.

63


whether such a structure could reasonably be described as targeting ‘higher’ earners and cause the cash flow issues described above.

3.35 When assessing the options for different treatment of people based on their level of earnings against the four principles, the Commission was considering:

• affordable and sustainable: whether the option shares risks optimally, the impact on the size and certainty of future benefit payments and the short-term implications for government cash flow;

• adequate and fair: whether the design would give all scheme members a good chance of reaching an adequate income in retirement and whether benefits are proportional to contributions paid;

• support productivity: the effect of each option on staff motivation, as well as how the scheme impacts on recruitment and retention; and

• transparent and simple: whether the scheme is understandable and allows members to plan for their retirement.

64


Table 3.E: Assessment of different scheme designs for higher earners against the principles

Affordability and sustainability

Adequate and fair Transparent and simple

Supporting productivity

Self-provision above a cap

Greatly reduces risk borne by Govt. Short-term cash flow problems but lower benefits paid out in future which increases sustainability

Core level of benefit ensures minimum income level for all, but higher earners may not reach desired replacement rate if they do not make own provision

Reasonably simple to understand. Higher earners may need information about how to make their own provision.

Reduces incentive for higher earners to stay in or join public services.

Employer-provided funded DC provision above a cap

Greatly reduces risk to Government. Loss of contributions due to funded DC provision would cause cash flow issues.

Core level of benefit ensures minimum income level for all, but higher earners may not reach desired replacement rate if DC scheme performs poorly

Potential for confusion on total level of benefits being provided

Provides decent incentives for higher earners to stay in or join public services

Less generous DB provision above a cap

Reduces scale of risk to Govt. Contributions can be collected on all earnings and future benefits to be paid out are reduced

Core level of benefit ensures minimum income level for all, but higher earners may not reach desired replacement rate

Different accrual rates could cause confusion for scheme members

Provides decent incentives for higher earners to stay in or join public services

A single scheme across all income levels

No reduction in risk for Govt, but contributions can be collected on all earnings

Ensures adequate levels of income. Higher earners may benefit disproportionately from schemes

Easily understandable

Provides good incentives for higher earners to stay in or join public services. All members are in the same scheme

Key: Poor outcome Fair outcome Good outcome


65


3.36 Although higher earners are better positioned than lower earners to bear some of the risks associated with pension provision, the introduction of a cap on pensionable earnings or a hybrid scheme does not seem attractive to the Commission due to:

• increased complexity, in terms of administration and ease of understanding;

• the cash flow implications caused by the loss of contributions on earnings above the cap;

• the minimal impact that the introduction of a cap at a reasonable level will have on the improvement in the level of benefits offered below the cap;

• the impact on recruitment and retention of higher earners; and

• the Commission’s view that there are some benefits for senior managers to be in the same scheme as their staff.

Recommendation: A single benefit design should apply across the whole income range (Recommendation 9a).

Indexation

3.37 The exact level of future pension payments from DB pension schemes is uncertain. This uncertainty stems from a number of factors, one of which is how much accrued pension benefits are increased annually (the level of ‘indexation’), both before and after retirement.

3.38 Indexation is the revaluation of pension benefits in line with some specified index, or according to some defined formula. The main aim of indexation is to maintain the value of the pension benefits over time.

3.39 Currently, public service pension scheme members are protected from the risks of price inflation after retirement, as increases to pensions in payment are linked to the statutory pension increases for State Second Pension. Government, and ultimately the taxpayer, therefore bears the risk of increasing prices.

3.40 The indexation of pre-retirement benefits is provided through the link to final pensionable salary in the final salary design. Some risk is shared with members as an individual’s pay is not guaranteed to rise in real terms. However, over long periods pay has tended to rise considerably faster than prices (actuaries within government usually assume an underlying trend of general pay increases of 1.5 per cent per annum above the Retail Prices Index (RPI)).

66


3.41 In current public service career average schemes, a number of different pre-retirement indexation methods have been adopted. The civil service Nuvos scheme links to the statutory pension increases for State Second Pension and public service pensions in payment, which from April 2011 will be the Consumer Prices Index (CPI). The NHS career average designs for GPs and dentists used to link to average increases in GPs’ or dentists’ earnings or profits, and now use RPI+1.5 per cent instead as a proxy for long-term growth in average earnings.

Post-retirement indexation

3.42 The desire to provide adequate pensions means that it is important to maintain the real value of members’ accrued pension benefits. Post-retirement, it is maintaining the purchasing power of the pension that is important.

Recommendation: Post-retirement, pensions in payment should be indexed in line with prices to maintain their purchasing power and adequacy during retirement (Recommendation 8b).

The options for pre-retirement indexation

3.43 The Commission considered a range of issues in relation to the indexation of benefits pre-retirement, specifically:

• the trade off between pre-retirement indexation and accrual rate, and how this distributes benefits amongst scheme members with different salary and service profiles;

• whether pre-retirement indexation should be in line with prices or some measure of earnings; and

• whether pre-retirement indexation should be the same for current (active) and former (deferred) public service scheme members who have yet to retire.

The trade-off between indexation and accrual rate

3.44 There is a trade-off between the level of indexation and the accrual rate. A low indexation and high accrual rate scheme could be designed that is expected to provide a similar level of benefits over a full career as a high indexation and low accrual rate scheme.

3.45 For a given cost, accrual rates will be more generous if a less generous indexation method is adopted, and vice versa. For example, earnings indexation will generally be more generous than prices indexation (since earnings typically outpace inflation over the medium to long term) and so the accrual rate will need to be lower in a CARE scheme with earnings

67


indexation. Affordability for the new scheme design can therefore be achieved by the balancing of indexation method and accrual rates.

3.46 Modelling conducted by the PPI6 has determined that the three benefit designs shown in Table 3.F are, on average, of equivalent expected average value to scheme members.

Table 3.F: Scheme designs of equivalent expected average value to members

Accrual rate Pre-retirement indexation (actives)

Pre-retirement indexation (deferreds)

Post-retirement indexation

Proxy for the current final salary schemes

1 / 60ths Final salary link Prices Prices

CARE scheme with higher accrual / lower indexation

1 / 40ths Prices Prices Prices

CARE scheme with lower accrual / higher indexation

1 / 61sts Earnings Earnings Prices

Source: Pensions Policy Institute.

3.47 The modelling takes a broad brush approach to determining schemes of equivalent value, and so the results should be viewed as approximate. Full costing calculations based on full scheme data would be needed to determine the exact parameters that would lead to equivalent expected value. However, the modelling gives a reasonable approximation to the parameters.

3.48 The choice of indexation is therefore not solely related to cost, but also has an impact on both adequacy and fairness. The combination of accrual rate and indexation method has a direct impact on how valuable each year of accrued service is to the scheme member. It therefore determines the distribution of the value of benefits between members of different ages and with different career profiles. If indexation is in line with prices, then service accrued near retirement is the most valuable, and service accrued earlier in a career has progressively less value the further back in time it was accrued. If indexation is in line with earnings, then service accrued early in the member’s career retains more of its value (in terms of proportion of salary at retirement) relative to the final year of accrual before retirement.

3.49 As an illustrative exercise, Chart 3.B below shows the value (pension accrued in terms of proportion of salary at retirement) to a hypothetical member of a year of accrual at each age from ages 25 to 65. The calculations are based on the two CARE schemes detailed in Table 3.F.

6 Further details of the modelling are given in Annex C.

68


Chart 3.B: The value of each year of accrued service in a CARE scheme

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

25 30 35 40 45 50 55 60 65

Prop

ortio

n of

sal

ary

at re

tirem

ent

accr

ued

AgeCARE with earnings indexation CARE with prices indexation


Note: Salary increases throughout the member’s career are assumed to be in line with the growth in average earnings. If

instead the member’s salary increases were higher than the growth in average earnings, then both lines would slope more

steeply upwards from left to right, and vice versa.

3.50 The analysis assumes the same level of indexation for active and deferred members. It is worth bearing in mind however, that if the scheme design had deferred indexation at a lower level than active indexation then benefits would be worth less to those who leave, and the value of each year of accrual would then also depend on the number of years spent as a deferred member of the scheme.

3.51 Chart 3.B shows that in a high accrual and prices indexation CARE scheme, one year of pension accrual at age 64 is worth almost two and a half times as much as one year of pension accrual at age 25.

3.52 A high accrual rate and prices indexation scheme therefore benefits those members who accrue service later in their careers, and is less valuable to those who accrue service earlier in their careers. The balance of indexation level and accrual rate will therefore have a direct impact on fairness between scheme members; for example, in a high accrual and prices indexation CARE scheme a public servant who works five years in the public sector at the start of their career will receive significantly less than a public servant who works five years in the public sector at the end of their career. At the extreme, the scheme could offer so little value for a young member relative to the employee contributions paid, that it would be rational for them to opt out as they could receive a better return on their contributions from a different savings vehicle.

69


3.53 The age disparity is much less for a low accrual and earnings indexation CARE scheme. One year of pension accrual is worth the same at all ages, since salary increases and pre-retirement indexation are both assumed to be in line with average earnings.7

3.54 Low accrual rates and earnings indexation is therefore more equitable between younger and older scheme members, and between members who work the same number of years in public service, but at different points in their career. Chart 3.C illustrates this, and shows replacement rates at retirement for low flyers8 under some simplified career path structures in public service.

Chart 3.C: Illustrative replacement rates at retirement for a low flyer

0%

10%

20%

30%

40%

50%

60%

70%

80%

CARE - earnings indexation/low accrual rate CARE - prices indexation/high accrual rate

Rep

lace

men

t rat

e

Long career (40 years)Short career early in working life (20 years) Short career later in working life (20 years)


Note: Indexation is assumed to be at the same level for both active and deferred members.

3.55 Chart 3.D shows that the results are similar for high flyers albeit that there is greater disparity in outcomes between a prices and earnings indexation scheme for high flyers compared to low flyers. This is due to the higher salary growth assumed for high flyers.

7 In the case of a member who receives salary increases in excess of average earnings, the low accrual / earnings

indexation CARE scheme would show that the value of accrual is not flat, but increases with age. However, under such a

scheme the age disparity would still be less than under a high accrual / prices indexation CARE scheme.

8 Further details of the modelling are given in Annex C, including a definition of low and high flyers.

70


Chart 3.D: Illustrative replacement rates at retirement for a high flyer

0%

10%

20%

30%

40%

50%

60%

70%

80%

CARE - earnings indexation/low accrual rate CARE - prices indexation/high accrual rate

Rep

lace

men

t rat

e

Long career (40 years)Short career early in working life (20 years) Short career later in working life (20 years)


Note: Indexation is assumed to be at the same level for both active and deferred members.

3.56 It can be seen that the CARE scheme with earnings indexation and a low accrual rate produces the least disparity in the value of pension benefits between members who work a similar number of years in public service, but at different points in their careers. A CARE scheme with earnings indexation and a low accrual rate will not penalise those members who work in public service early in their working lives, before leaving the public sector. A scheme such as this will therefore be of particular advantage to those who have caring responsibilities later on in their adult life, for example women who may work in public service early in their working lives before leaving the labour market to raise a family. It is worth noting that if the scheme design had deferred indexation at a lower level than active indexation then the replacement rates for the short career early in working life scenario would be lower than those shown in Charts 3.C and 3.D.

Pre-retirement indexation: prices or earnings

3.57 As set out in Chapter 2, the Turner Commission outlined a set of benchmark replacement rates that specified an adequate retirement income in terms of an individual’s income at retirement. Therefore some link between earnings and the pre-retirement indexation method may be appropriate, in order that benefits accrued early in a career maintain their value relative to the wealth of the working population and to the individual concerned.

71


3.58 The adequacy principle suggests that there is an argument for earnings indexation pre-retirement, since the value of the accrued pension benefits would be maintained relative to the wealth of the working population. Earnings indexation also scores well against the fairness principle, as it leads to more equitable outcomes between members of different ages.

Recommendation: Pension benefits should be uprated in line with average earnings during the accrual phase for active scheme members (Recommendation 8a).

Indexation of deferred benefits

3.59 The Commission’s interim report stated that “labour mobility supports an efficient labour market and enhances productivity. Pension scheme design should not be an unintended barrier to movement into, or out of, the public sector.”9 The final salary structure of the current public service pension schemes discourages movement of individuals from the public to the private sector due to two factors:

• the lower level of indexation applied to leavers’ benefits (deferred pensions); and

• the increase in benefits resulting from promotions later in an individual’s career.

3.60 The removal of the final salary link would eliminate the barrier due to the second factor. However if different levels of indexation were to be applied to active and deferred members (earnings for active members, prices for deferred members, for example) then the barrier due to the first factor would remain. Differing levels of active and deferred indexation also raises the issue of fairness, since this would mean that a year of accrual would be worth different amounts to a member at retirement depending on whether they remain in or leave public service employment. This would have implications for adequacy, as accrued benefits may not maintain their value relative to the wealth of the working population.

3.61 Adopting the same level of pre-retirement indexation for both active and deferred members would in principle aid labour mobility as there would then be no financial detriment to the member when leaving public service employment, aiding transition of employees from the public to the private sector. This would also achieve fairness between scheme members who work in public service for a similar number of years, but at a different point in their careers. The actual impact on the decision to move sectors will depend on the individual’s understanding and valuation of how their pension will be affected.

3.62 Alternatively, if the Government assesses that an important role of public service pensions is to encourage staff retention, then pre-retirement indexation for active members could be set at a higher level than that for deferred members. For a given cost, this would also mean that a higher accrual rate could be offered to active members, as deferred members

9 IPSPC (2010) Interim Report, p.102.

72


would benefit less from the scheme. It should be noted that if this strategy is adopted then mobility and fairness issues would remain, albeit to a lesser extent than under current final salary arrangements.

3.63 The greater the variation in indexation before and after retirement, or between active and deferred members, the more complicated the scheme would be to administer and for members to understand. The principles of simplicity and transparency therefore suggest that variation should be kept to a minimum, unless there is a specific need for variation to meet an objective.

3.64 The choice of either prices or earnings indexation for deferred members should depend on the Government’s assessment of the role of public service pensions and the balance between the competing principles of adequacy and fairness, and supporting productivity.

Caps on indexation

3.65 The sharing of risk between scheme member and government could be adjusted by introducing caps on indexation. For example, some of the inflation risk borne by government under a prices indexation methodology could be transferred to the members by capping increases. In the private sector only around 1 in 6 active members of DB schemes are guaranteed uncapped inflation protection,10 so it could be perceived as unfair that public service pension scheme members are fully protected from any increases in prices.

3.66 However the decision to introduce a cap on indexation should not be taken lightly. It presents a very real and significant risk to members during periods of high inflation, as the real value of their benefits could be significantly reduced over a relatively short period of time. The Commission is of the view that government is better placed to bear inflation risk than scheme members, and so a cap should not be introduced. This will provide certainty at the individual level, while the robust control mechanisms that the Commission is recommending (see Chapter 4) can be used to control costs.

3.67 The issue of reductions to pension benefits, to take account of periods of deflation for example, should be considered. If there is no mechanism for reducing pensions in payment to maintain their real value then there is asymmetric sharing of risk between members and government, since government bears the risk of high inflation and members benefit from periods of deflation. This leads to unfairness between members and the taxpayer. Reducing pensions in payment would be challenging for pensioner members, and so a cumulative indexation regime should be considered, whereby negative adjustments during periods of deflation can be carried forward to reduce positive adjustments in future years.

10 Occupational Pension Scheme Survey Annual Report 2009, Table 4.14, Office for National Statistics.

73


The impact of the proposed system on public service pension scheme members 3.68 To assess the impact on adequacy of different possible scheme designs, the Commission has carried out modelling (described in Annex C) to see how the pension incomes of future public service employees could be affected by the three principal schemes modelled by the Pensions Policy Institute outlined in Table 3.F.

3.69 The three schemes perform very similarly in achieving adequacy targets. About two thirds of people who work in the public sector at any point in their careers would receive adequate pensions (as defined by the Turner Commission) at their State Pension Age.

3.70 Long-serving public sector employees are very likely to receive adequate retirement incomes from sources including public service pensions, state pensions and private sector pensions. Chart 3.E looks at the impact of a CARE scheme with indexation by average earnings on employees with at least 20 years of service in the public sector. Adequacy targets are achieved for more than 90 per cent of those who retire in one of the bottom three income brackets, and for more than three quarters of those in the highest income bracket.11 There is little variation between the different scheme designs in this respect, with the final salary proxy scheme and CARE with inflation indexation also performing well.

Chart 3.E: Proportion of members with careers longer than 20 years achieving adequacy target in a CARE scheme with earnings indexation

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Less than £9,500 £9,500-£17,499 £17,500-£24,999 £25,000 and above

Perc

enta

ge o

f mem

bers

Gross income band on retirement


11 The gross income bands are based on earnings levels in 2005. The Commission’s analysis adjusts these bands in line

with average earnings growth.

74


3.71 There are though important differences in how the different schemes distribute their pension payments. As discussed in Chapter 1, final salary schemes are disproportionately beneficial to high flyers. This is reflected in Chart 3.F, which demonstrates that the lowest tenth of pension payments are on average almost 90 per cent higher in a CARE scheme with indexation by average earnings than in a final salary scheme, and a third higher in a CARE scheme with indexation by inflation. The top tenth of pension payments is about 6 per cent lower in a CARE scheme with indexation by average earnings, and 4 per cent lower in a CARE scheme with indexation by inflation.12

3.72 Relative to a final salary design, a CARE scheme would therefore redistribute pension payments from the highest earners to those in lower deciles.

Chart 3.F: Pension payments from CARE schemes relative to final salary scheme, by deciles of pension income

0%20%40%60%80%

100%120%140%160%180%200%

1 2 3 4 5 6 7 8 9 10

Decile

Earnings indexation Inflation indexation


3.73 This analysis shows that there is no reason for a move away from final salary pensions to result in inadequate retirement incomes, but that there are likely to be winners and losers from a switch.

12 Because payments in the top decile are on average much larger than those in the bottom decile, much smaller

percentage differences are associated with similar differences in total expenditure.

75


Contributions and opt-out rates

3.74 Employee contribution rates across public service schemes vary considerably, both in level and structure. As a result public service workers in different pension schemes, and in some cases within the same pension scheme, pay a different proportion towards the cost of their pensions.

Table 3.G: Current contribution levels

Scheme Employee contribution (%)

Employer contribution (%)

Proportion of total contributions paid by employees (%)

Teachers 6.4 14.1 31

NHS 5.5 - 8.5 14 26 - 38

Civil service 1.5 or 3.5 19a 6 - 17

Police - 1987 scheme 11 24.2 31

Police - 2006 scheme 9.5 24.2 28

Fire - 1992 scheme 11 26.5 29

Fire - 2006 scheme 8.5 14.2 37

Armed Forces - officers 0 37.3 0

Armed Forces - others 0 21.4 0

LGPS 5.5 - 7.5 14 - 25b 18 - 35

Average private sector (open DB)c

5.4 14.9 27

Average private sector (open DC)c

3.0 6.4 32


a) Average rate, employer contribution rates tiered by earnings.

b) Employer rates determined at individual fund valuations, and include contributions to fund past deficits, as do many of

the employer contributions to the unfunded schemes.

c) Occupational pension schemes survey 2009.

3.75 In the 2006 Fire Pension Scheme, employees pay 8.5 per cent of pay – almost 40 per cent of the total contribution required with the employer picking up the remaining 60 per cent. In the civil service, some longer serving employees pay 1.5 per cent – closer to 10 per

76


cent of the total contributions required with the employer therefore paying over 90 per cent.13

3.76 The current disparity in contribution levels across the schemes is the result of historic developments and negotiations over a number of years at the scheme-specific level. Therefore the rates will reflect pay negotiations and differing pension benefit structures, and as such cannot be considered independently of overall remuneration.

3.77 In reviewing contribution rates the Commission considered four key questions:

• what factors should influence the appropriate split of contributions between employer and employee?

• should contribution rates vary by member?

• what is the impact on opt-out rates? and

• should contribution rates be more uniform across public service and if so, what factors should be considered during transition?

Factors influencing the split of contributions

3.78 Theoretically, for scheme members it should not be the level of employee contribution that matters but the overall compensation package. In simple terms, for example, individuals should be indifferent between receiving £80 today plus an employer contribution to their pension of £20 or £90 today and having to make employee contributions of £10 of this for their retirement themselves, in addition to an employer pension contribution of £10. In practice, however, the level of employee contributions does matter. Asking employees to pay a contribution at all increases awareness that providing them with a pension costs money. Determining what the level of contributions should be is then a balance between competing factors.

3.79 High contribution rates could discourage participation, either because lower income employees are financially constrained and consider that they cannot afford to make higher contributions, or because people irrationally underestimate the value of the benefit being provided because it will occur so far into the future. Low employee contributions incentivise participation in the scheme due to the larger employer contribution which the employee would have to forgo if they opted out. Employees choosing to opt out rather than pay the pension contribution, particularly for those employees on low wages, has implications for adequacy of provision in retirement and potentially on means-tested benefits expenditure.

13 These total contribution rates are based on the SCAPE discount rate of RPI + 3.5 per cent, which the interim report stated was at the high

end of what is considered reasonable. Using a lower discount rate would result in higher total contributions required, and therefore on the

current rates, the employee proportion would be lower.

77


3.80 However, maintaining rates at a low level in the face of increasing pension costs will mean that employers will need to contribute more, which is costly and could lead to the schemes becoming unsustainable. This may require reductions in the generosity of pensions, which in turn could threaten adequate income levels in retirement. Rates very different from those seen in the private sector may be a barrier to labour market mobility.

3.81 Therefore contribution rates should be set so that members appreciate the value of their pension (so schemes should be contributory). Care needs to be taken so that contributions are not so high that they lead to scheme members, especially at lower income levels, opting out of the pension scheme.

Variations by member

3.82 The total contribution rate required from employers and employees to cover the cost of benefits accruing in a scheme is a single rate determined on a scheme wide basis. It is effectively a weighted average of the rates applicable to each member. Across a range of employees paying the same contribution rate, there are numerous cross-subsidies between members with different characteristics, meaning some members receive much better value from a fixed contribution rate than others. The degree of these cross-subsidies will depend on the scheme structure being considered.

3.83 Some public service pension schemes, for example the NHS and local government schemes, have contribution rates tiered by earnings. Tiered contributions were introduced to reflect the fact that high-flying high earners receive proportionately more benefit from the current final salary arrangements than lower earners, per pound of employee contributions paid.

3.84 Once the scheme design changes from final salary to CARE or cash balance this justification becomes less valid, since the discrepancy between the benefit derived from the scheme for high and low flyers is much reduced due to the loss of the final salary link.

3.85 However there are other reasons why tiered contributions would remain appropriate. In general higher earners have a higher life expectancy and so may receive a pension for a longer time than those with lower earnings. Chart 3.G shows that individuals with higher pensions (which are a proxy for higher earnings and wealth) experience lower mortality than individuals with lower pensions. It is therefore reasonable to ask higher earners to contribute more, as they are likely to get more benefit from the scheme due to their pension being paid for longer.

78


Chart 3.G: Relationship between pension size and male mortality in self-administered pension schemes

020406080

100120140160180200

50-54 55-59 60-64 65-69 70-74 75-79 80-84 85-89 90-94

Perc

enta

ge o

f rel

ativ

e m

orta

lity

Age

£0 - £2,999 £3,000 - £4,499 £4,500 - £8,499

£8,500 - £12,999 £13,000+

Source: CMI Working Paper 31.

Note: Actual mortality rates of male pensioners in the SAPS study shown against that expected on the basis of the

PCMA00 series tables. SAPS study for period 2000-06 relative to ‘00’ series mortality.

3.86 Furthermore, Chart 3.H illustrates how participation rates vary across all public service pension schemes by salary and by employment status. The general trend shows increasing participation with increasing salary, flattening out at higher earnings levels. The opt-out rate for those earning under £18,000 is almost 20 per cent, increasing to over 30 per cent for part-time workers.

79


Chart 3.H: Participation rate by earnings

60%65%70%75%80%85%90%95%

100%

Parti

cipa

tion

rate

Part-time workers Full-time workers

Source: IPSPC analysis of data from the Annual Survey of Household Earnings.

3.87 There is therefore a case for different treatment of higher and lower earners to encourage greater participation amongst low-paid and part-time workers. It could be argued that contributions should be weighted so that those who earn more pay the most. Lower contribution rates for the lower paid encourages participation in pension schemes, contributing to higher provision in retirement and less reliance on state benefits.

3.88 Since a cap on pensionable pay or a hybrid scheme have been ruled out by the Commission, tiered contributions are also an effective way of differentiating between higher and lower earners. In the absence of a cap higher earners are able to accrue disproportionately higher pension benefits than lower earners, and so benefit more from the guarantees that government offers through the scheme. It is therefore reasonable to ask higher earners to contribute more to the cost of the scheme for the higher level of risk the scheme has underwritten.

Recommendation: The differing characteristics of higher and lower earners should be addressed through tiered contribution rates (Recommendation 9b).

Uniformity, transition and the impact on opt-outs

3.89 Given the current range of rates and structures any transition to a more uniform structure for employee contribution rates will be complex. As mentioned previously, employee contribution rates fit into an overall remuneration package. In some schemes low rates may be matched by lower pay. At the extreme, the armed forces pay no contributions

80


but an allowance for this is recognised when considering overall remuneration. Arguably, adjustments could be made to salaries to remove such effects across schemes without impacting on take-home pay to the employees, or to the cost of employment. However in order to do this, an assessment would be needed of how much of the difference in contribution rates between schemes is as a result of pay effects. The implications for total renumeration extend beyond the scope of this report and may require separate consideration.

3.90 Any increase in contribution rates is likely to result in some increase in opt-out rates. Even if met by an increase in salary, the extra cash available from opting out may entice employees out of the scheme. Managing the risk of opt out would be key to transitioning to a more uniform approach. At the very least, changes should be phased in over a specified period. Increasing education and awareness of the value and importance of pensions should also discourage opting out.

Recommendation: The Government should consider the trade off between affordability and the impact of opt outs on adequacy when setting member contribution levels (Recommendation 9c).

Box 3.F: Changes to contracting out of State Second Pension

Arrangements for contracting out of the State Second Pension will be abolished for all those in defined contribution schemes from April 2012 and it will also become a flat-rate top-up to the basic State Pension (bSP) by 2030 for those contracted in. This will mean that from 2012 contracting out will only be available to members of defined benefit schemes. The vast majority of these members are in public service schemes.

The move to a flat-rate system means that National Insurance contributions will gradually rise for those who are still members of contracted out schemes and are earning above £40,040 per annum, the majority of whom are in public service defined benefit schemes.

This increase in national insurance contributions, as well as the impact of any future reform of the state pension system, will need to be taken into consideration by the Government when setting contribution rates in the reformed public service schemes. It should be mindful of the potential effect on opt-out rates.

Choice

3.91 Individual choice and flexibility should be important parts of the pension scheme landscape. They can bring benefits for scheme members and government alike, by enabling scheme members to tailor their pension provision to their requirements. Many current public service pension schemes provide significant choice in some areas, for instance of the age at which pension payments begin and whether to take a retirement lump sum.

81


3.92 Although choice empowers scheme members, there are risks involved. In particular, individual choice can lead to decisions that may harm either the member or possibly society as a whole. Limited public information, for instance about the risks involved in choosing different pension products, can lead members to make poor decisions. Even with good information, the complexity of pensions can result in poor decisions. Short-sighted decision making could leave the employee regretting in later life that they had saved too little.14

3.93 Even if employees make decisions that are individually rational, they might not be socially beneficial. Implicit guarantees against pensioner poverty could cause people to undersave, expecting to be bailed out in the future. Early retirement might be good for the employee but bad for society, for instance if continued employment would generate benefits to others (perhaps by increasing the productivity of other workers). Adverse selection (whereby, for instance, only those who expect to live longest choose to buy annuities) can cause markets to unravel and make almost everyone worse off.

3.94 There can also be significant potential administrative costs in providing for employee choice. There are typically economies of scale in both administration and fund management, implying that the cost per worker will be lower if everyone is provided with the same scheme design and management. Moreover, it could be costly to maintain the infrastructure necessary to enable employee choice, including advice services and implementation.

3.95 Choice should therefore be optimised, not simply maximised. Faced with too much choice, people can become overwhelmed and disengaged;15 Chart 3.I shows that fewer than half of men, and a third of women, believe that they understand enough to make a decision about retirement saving. In some areas, increasing choice would lead to an unacceptable risk of members failing to achieve adequate pension incomes in retirement. Pension schemes should therefore pick the most valuable areas in which to offer choice, bearing in mind the following:

• keep choices simple;

• use automatic enrolment;

• design a good default option; and

• allow people to commit now to (reversible) action in the future, thus using procrastination to assist policy.16

14 Hyperbolic Discount Functions, Undersaving, and Savings Policy, Laibson, 1996.

15 Choice overload and simplicity seeking, Iyengar and Kamenica, 2006.

16 Reforming pensions: principles, analytical errors and policy directions, p.12, Barr and Diamond, 2009.

82


Chart 3.I: Percentage of respondents who understand enough about pensions to make a decision about saving for retirementa in 2006-08

0

10

20

30

40

50

60

70

16 to 24 25 to 34 35 to 44 45 to 54 55 to 64 All

Perc

enta

ge o

f res

pond

ents

Age group

Men Women

Source: ONS, Wealth and Assets Survey.

a) Individuals below SPA in Great Britain and not retired.

3.96 The Commission’s analysis suggests that most beneficial choice is connected to pension age and additional pension contributions, within a strong core pension system that will provide an adequate retirement income for those with full public service careers.

3.97 Members already have significant choice over when to start receiving their pension benefits, above the minimum pension age. But there are concerns that it is currently difficult to make informed choices, and members may not understand the trade-offs involved in choosing an early or late retirement age. Chapter 6 discusses ways in which pension schemes can be made more transparent and easier to understand.

3.98 Actuarial reduction means that someone who retires early draws a reduced pension that reflects the greater cost of paying a pension for longer, and the risk of death in the intervening period. Similarly actuarial enhancement means that someone who retires late draws an increased pension that reflects that the pension will be paid out for a shorter time. Actuarial reduction and enhancement therefore adjust pensions on a cost neutral basis: there are no subsidies for early retirement and no incentives against continuing to work and retiring later. There are strong arguments for ensuring actuarial enhancement and reduction for future public service pensions. With increased longevity, it is particularly important not to penalise longer working lives.

3.99 For similar reasons, caps on total pension accrual (for example, a limit on the maximum number of years of pensionable service) should be removed or significantly lifted so as not to discourage people from having a longer working life.

83


3.100 Replacing or removing abatement of pensions and increasing or removing caps on pension accrual will have cost implications for pension schemes which will need to be seen as part of the wider package.

3.101 Flexible retirement, whereby employees continue to work while drawing some proportion of their pension, is likely to be increasingly important as working lives increase. As well as personal preferences, disability, ill health and family commitments could all lead employees to prefer to reduce their workloads in their 50s or 60s. Both the post-reform NHS scheme and the civil service nuvos scheme have introduced flexible retirement provisions. The Swedish state pension system is another possible model here; it allows members to draw 25 per cent, 50 per cent or 75 per cent of their pension from the minimum pension age, while the remaining proportion continues to increase. Employers would need to be receptive to part-time working by older workers, or moves to jobs with reduced responsibilities.

3.102 At present, most public service pension schemes provide for abatement, where some people who return to work in a job covered by the same scheme from which they draw their pension receive a reduced pension as a result. This can have a significant negative impact on an employee’s effective salary if he returns to work. The resulting disincentive for work could harm employees, employers and taxpayers, and looks hard to justify in the context of increasing working lives.

3.103 Abatement should therefore be eliminated under the new system, or replaced with a structure whereby pensions are initially reduced if an employee returns to work, but are then increased commensurately when the employee finally retires fully (as occurs in the US social security system).

Recommendation: Members should have greater choice over when to start drawing their pension benefits, so they can choose to retire earlier or later than their normal pension age and their pension would be adjusted accordingly on an actuarially fair basis. Flexible retirement should be encouraged and abatement of pensions in its current form for those who return to work after drawing their pensions should be eliminated. In addition, caps on pension accrual should be removed or significantly lifted (Recommendation 10).

3.104 There is good reason to enable additional pension contributions to help retirement planning. Members approaching retirement who feel that their pension will not be sufficient (perhaps because of short service) could thus choose either to make additional contributions or to work for longer.

3.105 The format of additional contributions should be clear and transparent to employees. Defined benefit added pensions, as are currently available in several schemes, are particularly valuable, as they mean that employees can be confident in the relationship between their contributions and future pension receipts. These should be made available on a fiscally neutral basis – any subsidy to additional contributions would tend to privilege wealthier employees, as they are more likely to take advantage of the opportunity to contribute more.

84


3.106 Lump sums are currently available to members on retirement through pension commutation (exchanging a certain proportion of pension for a lump sum) or as a default option. It is possible to take a tax-free lump sum up to a limit of 25 per cent of the pension value. Lump sums are popular with members and can provide valuable flexibility by giving members a buffer fund to deal with contingencies. Therefore lump sums should continue to be made available to members of public service pension schemes through commutation.

3.107 The Commission has also explored whether there should be choice around the level of ancillary benefits that are provided by the public service pension schemes. This is explored in more detail in the next section.

Ancillary benefits

3.108 Pension schemes in both public and private sectors typically offer a range of ancillary benefits, which are akin to insurance and are not directly linked to the pension itself. The main ancillary benefits are:

• death benefits, which may become payable on the death of an active scheme member (death-in-service), someone with preserved pension rights (death-in-deferment) or upon the death of a scheme pensioner (death-in-retirement);

• dependants’ benefits, which are payable to nominated partners (survivors’ benefits) and dependants (children’s benefits);

• pension guarantees, which guarantee to pay the member’s pension for a minimum period after retirement; and

• ill-health/injury-related benefits, which are triggered when the scheme’s medical adviser agrees that the member is unable to perform their current job.

3.109 Ancillary benefits (with the exception of dependants’ benefits) are used by a small proportion of public sector employees but this is not surprising, given that they are similar to insurance in nature. For example, death-in-service benefits are, happily, infrequently used17 and ill-health public sector retirement (estimated at 22,000 annually in the late 1990s)18 has fallen significantly in the last decade,19 due to measures to tighten access and encourage healthier workplaces.

17 Less than 0.5 per cent of active members die in service annually. For example, 210 active members of the Armed

Forces pension scheme died in 2009-10, out of an active membership of 198,032 (0.11 per cent); 368 out of 609,534

active members of the Teachers Pension Scheme (0.06 per cent) died in 2009-10.

18 Review of ill-health retirement in the public sector, HM Treasury, 2000.

19 For example, ill-health retirements in local government fell from 14.9 per cent of all retirements in 2004-05 to 7.4 per

cent in 2008-09; in the NHS, ill-health retirements fell from 23 per cent of all retirements in the late 1990s to 8 per cent in

2009-10.

85


3.110 Although these benefits are used by a small number of members, they have been highly regarded whenever government has consulted on possible changes to schemes. The available evidence also suggests that the costs of providing these benefits are relatively small.20

Options for future ancillary benefit provision

3.111 The Commission considered a number of options in terms of ancillary benefit provision – amending these benefits, offering more choice in their provision, removing them entirely, providing them through the private sector or retaining them in their current form.

3.112 There appears to be limited value from reducing or removing ancillary benefits. For individuals, any reductions in contributions would be marginal and are likely to be outweighed by the potential advantages of having these benefits (or the potential costs of not having them). For the Exchequer, any short-term savings may be outweighed by long-term costs. As mentioned in the previous section on choice, evidence suggests that people often choose inefficiently, underestimating their long-term needs and those of their families. Reducing or removing these benefits may result in people undersaving and thereby increasing their reliance on the State when they are older.

3.113 Offering more choice on ancillary benefits would require schemes to provide people with a lot of information and support in order to make informed decisions. Providing choice would also make the system more complex and expensive to administer, and would increase the risk of adverse selection. And as was mentioned in the previous section, the ideal scenario is for choice to be optimised, rather than maximised.

3.114 Although information on the costs of ancillary benefit provision appears limited, the Commission has seen nothing to suggest that the private sector is better placed than the public sector to provide these benefits for public service pension scheme members, or that they could do so more cheaply. Government should be able to obtain economies of scale that most private sector schemes could not, and private sector provision is also likely to contain a profit element and to price to reflect risk. Therefore, ancillary benefits should be broadly retained in their present form. The current system provides people with equal access to a range of valued, cost-effective benefits on top of their pension, which provides a useful safety net.

20 There is little specific evidence the Commission has seen on the cost of ancillary benefits. It is difficult to cost and

compare the individual elements of ancillary benefits between schemes; these benefits are inherently uncertain in nature

and scheme valuations take place at different times, sometimes with varying methodologies. Papers from the Government

Actuary’s Department indicate that ancillary benefits typically cost a few per cent of pensionable pay.

86


87


4 The controls

Box 4.A: Summary

• Life expectancy has increased dramatically in the last few decades and future changes are uncertain. The Commission believes it is fair for members to bear the risks and costs of changes in life expectancy.

• There are a number of different mechanisms which could be used to share longevity risk. Certainty about how such mechanisms will be managed is essential for scheme members and taxpayers. The Government should therefore look to link a member’s Normal Pension Age (NPA) for most schemes so that it is in line with their State Pension Age (SPA). This will reflect changes in longevity over the last few decades and the likely nature of changes in the future.

• The link to SPA should be regularly reviewed to make sure NPA remains in line with changes in life expectancy, but with a preference for keeping the two pension ages aligned.

• The mechanism discussed above, together with the design recommended in the previous chapter should achieve the Commission’s aim of passing appropriate risks to members, which in turn will provide increased stability of cost for government, and ultimately the taxpayer.

• However, the Commission believes there will still be a need for an overriding mechanism to ensure that public service pensions are affordable and sustainable. The Government should introduce additional mechanisms for controlling for future changes in longevity, or other costs, by introducing a fixed cost ceiling which, if breached, triggers a change to bring costs under the ceiling. This should be the subject of consultation between the Government and scheme members, with an automatic default change if agreement cannot be reached.

4.1 The analysis of the risks in pension schemes in Chapter 2 concludes that it is appropriate for some risks to be shared with scheme members. The most significant of these risks is longevity. This chapter outlines how scheme design can be structured to share risks effectively and maintain control of scheme costs.

88


Longevity

Longevity has increased

4.2 As discussed in the interim report, longevity has increased significantly in the past few decades and the life expectancy of members of public service pension schemes is no exception. Chart 1.A illustrates how life expectancies from age 60 in England have increased for men and women in the general population since the 1840s.

4.3 This general increase in life expectancy has led to increases in the amount of time a public service pension scheme member can expect to spend in retirement. Chart 4.A compares the expected proportion of adult life spent in retirement for members retiring from the NHS Pension Scheme at age 60 over the past 50 years. The chart shows that current pensioners retiring at 60 can expect to spend around 40 to 45 per cent of their adult lives in retirement, compared with around 30 per cent for pensioners in the 1950s. This increase has been partially offset by increasing the Normal Pension Age (NPA) to 65 for new entrants into the scheme, though most members in public service pension schemes have a NPA of 60 or less.

Chart 4.A: Proportion of adult life spent in retirement, based on NHS Pension Scheme

28%

29%

37%

41%

34%

4%

5%

3%

3%

3%

0% 10% 20% 30% 40% 50%

1955 NRA 60

1979 NRA 60

2004 NRA 60

2010 NRA 60

2010 NRA 65

Per cent

Males Females (additional)

Source: IPSPC analysis of GAD valuation reports.

4.4 Although greater longevity is a positive development, the unprecedented rise in life expectancy since the schemes were set up has meant that providing public service pensions has become significantly more expensive than was anticipated. This change has had a

89


profound effect on the sustainability of public service pensions but has not been the focus of informed public debate. The cost of pensions in 2004 in the NHS was a third higher than it would have been if assumptions about life expectancy were the same as those in 1955. Similar results could be calculated for the other unfunded schemes. Since employee contributions have risen little over the period this increase in the cost of making pension commitments has mostly been paid for by employers and taxpayers.

Further changes in longevity are uncertain

4.5 It is generally assumed that longevity will continue to increase in the future, but there is significant uncertainty about the scale of any future changes. Increases in life expectancy have historically been recognised in future projections but the rate of improvements has been consistently underestimated. This is illustrated in Chart 4.B.

Chart 4.B: Actual and projected period life expectancy at birth for UK males

66

68

70

72

74

76

78

80

82

84

1966 1976 1986 1996 2006 2016 2026

Life

exp

ecta

ncy

at b

irth

Year

Actual2008 based2004 based2002 based1998 based1992 based1991 based1989 based1985 based1977 based1971 based

Source: IPSPC analysis drawing on C Shaw, 2007 and ONS 2008 population projections.

4.6 The chart emphasises that it is not possible to be confident about what the longevity experience will be in the future. Because of this, it is how the uncertainty is measured and managed that is important within a pension system. The implication of increasing, but uncertain, life expectancy is that schemes should have mechanisms in place to control the associated costs.

90


The State Pension and longevity changes

4.7 This observed and projected increase in life expectancy has already been recognised within the State Pension system with the Pensions Act 2007 proposing an increased State Pension Age (SPA) of 68 from 2046 for both men and women with interim increases to 66 and 67. Further details are set out in Box 4.B.

4.8 Given the continuing increases in longevity, the current Government last year reviewed the timetable for increasing the SPA, specifically the increase to age 66, with a view to accelerating this timetable. The Government has now decided to bring forward the move to age 66 from 2026 to 2020. Chart 4.C below shows the current timetable for increases in the SPA.

Box 4.B: Changes to State Pension Age

• Under the Pensions Act 1995, the SPA for women was to be equalised with that for men, rising from 60 in 2010 to 65 by 2020. Under the Pensions Act 2007, the SPA for both men and women was expected to rise to 66 by 2026, to 67 by 2036, and to 68 by 2046.

• However, official projections of average life expectancy have since been revised upwards. Life expectancy projections made in 2009 indicate that men and women reaching 66 in 2026 are expected to live, on average, at least 1.5 years longer than was thought at the time the Pensions Act 2007 was passed.a

• As a result, in June 2010 the Government issued a Call for Evidence on the timing of the increase in SPA to 66. This was followed in November 2010 by the publication of a White Paper outlining the Government’s decision.b Provisions to enable this were included in the 2011 Pensions Bill which was presented before Parliament in January 2011.

• Under the Bill’s new proposals, the equalisation of SPA between men and women, by which both men and women will have a SPA of 65, is planned to happen by 2018 rather than 2020. The SPA for men and women will then rise from 65 to 66 between December 2018 and April 2020, brought forward from 2026.

• The 2011 Pensions Bill makes no proposals for adjusting the timetabled rise of the SPA to 67 and 68. However, the Government signalled in the 2010 Spending Review that they are considering what future changes to SPA may be necessary.

a) DWP analysis based on Office for National Statistics Cohort Life Expectancy principal projections, for average life

expectancy for men and women resident in the UK.

b) DWP (2010).A sustainable State Pension: when the State Pension age will increase to 66.

91


Chart 4.C: Timetable for the changing State Pension Age

56

58

60

62

64

66

68

70

1945 (66) 1950 (61) 1955 (56) 1960 (51) 1965 (46) 1970 (41) 1975 (36) 1980 (31) 1985 (26)

Stat

e P

ensi

on A

ge

Year of birth (age in 2011)

Male SPA Female SPA Male SPA (revised) Female SPA (revised)

Source: IPSPC analysis of data from DirectGov.

Managing longevity changes in public service pension schemes

4.9 It is in the light of these changes in longevity and the SPA that the Commission considered the options for managing longevity risk within the public service pension schemes. As discussed in Chapter 2 the Commission’s view is that it is fair for the member to bear the longevity risk pre-retirement, or at least until close to retirement, as the Commission recognises the need for greater certainty closer to retirement. The Commission is also clear that bearing this risk means that tools to manage longevity should be activated if life expectancy decreases, and reduces scheme costs, as well as if life expectancy increases. The options considered and their assessment against the Commission’s principles is discussed below and set out in Table 4.A.

4.10 The Commission considered five options of managing longevity risk:

• the baseline (current scenario in most of the schemes for new members), setting the NPA at 65;

• conditional indexation pre-retirement linked to changes in longevity. How accrued benefits were uprated would then be dependent on changes to longevity;

• link a member’s NPA to the current SPA timetable. Members would only be able to draw down their pensions without actuarial reduction at their SPA, and this date would be set a number of years before retirement;

92


• link a member’s NPA to the current SPA timetable but with regular review of that link. This would work as above but with the addition of a regular review of the link to SPA to examine whether it was still appropriate given changes in longevity and other factors; and

• NPA automatically adjusted in line with set parameters, for example, it could be adjusted in line with life expectancy projections with the aim of targeting a set percentage or set number of years in retirement.

4.11 In all cases, the linking of NPA to SPA would apply to future service only. The protection of accrued rights means that the benefits built up within the existing schemes would still be payable at the current NPA in those schemes.

4.12 In Table 4.A these options are assessed briefly against the principles. When considering them against the principles the Commission was specifically considering:

• affordability and sustainability: whether the option manages past and projected increases in cost due to changes in longevity and whether it can manage the uncertainty surrounding future longevity;

• adequate and fair: whether the option is likely to deliver an adequate pension so it does not rely on decreasing annual pension benefit levels in the scheme to cope with changes, and whether it treats members from different cohorts in a similar way and could provide sufficient lead-in time for any changes;

• supports productivity: whether the option supports a changing length of working life in response to changes in longevity; and

• transparent and simple: whether the way the option worked would be easy to understand and allow members to plan for their retirement.

93

Independent Public Service Pensions Commission: Final Report - 10 March 2011Ta

ble

4.A

: A

ssess

men

t o

f lo

ng

evit

y m

an

ag

em

en

t o

pti

on

s ag

ain

st t

he p

rin

cip

les

Op

tio

ns

Aff

ord

able

an

d s

ust

ain

able

Ad

equ

ate

and

fai

rSu

pp

ort

pro

du

ctiv

ity

Tran

spar

ent

and

sim

ple

NPA

set

at

65

(bas

elin

e –

curr

ent

scen

ario

in

mos

t ne

w

sche

mes

)

Doe

s no

t m

anag

e th

e un

cert

aint

y ar

ound

long

evity

or

the

curr

ent

proj

ecte

d in

crea

ses.

No

man

agem

ent

of t

he lo

ngev

ity r

isk

thro

ugh

the

NPA

, so

cost

cha

nges

wou

ld n

eed

to b

e m

anag

ed

thro

ugh

othe

r m

echa

nism

s, w

hich

cou

ld r

educ

e ad

equa

cy. M

embe

rs f

rom

diff

eren

t co

hort

s, w

ith

diff

eren

t pr

ojec

ted

life

expe

ctan

cies

, wou

ld h

ave

the

sam

e N

PA.

No

cultu

ral e

xpec

tatio

n fo

r co

ntin

ued

wor

king

bey

ond

NPA

, eve

n th

ough

the

SPA

w

ould

be

incr

easi

ng.

Easy

for

mem

bers

to

unde

rsta

nd.

Con

ditio

nal

inde

xatio

n pr

e-re

tirem

ent

linke

d to

cha

nges

in

long

evity

Man

ages

mos

t of

the

pas

t an

d pr

ojec

ted

cost

s as

soci

ated

with

ob

serv

ed a

nd p

roje

cted

incr

ease

d in

lif

e ex

pect

ancy

. Aut

omat

ic a

djus

tmen

t of

the

inde

xatio

n m

easu

re w

ould

in

trod

uce

a m

echa

nism

for

dea

ling

with

unc

erta

inty

sur

roun

ding

lo

ngev

ity.

Inde

xatio

n w

ould

var

y in

acc

orda

nce

with

cha

nges

in

long

evity

, If

proj

ecte

d lif

e ex

pect

ancy

incr

ease

d th

is c

ould

lead

to

a re

duct

ion

in b

enefi

t le

vel a

nd

ther

efor

e ad

equa

cy.

Long

evity

cha

nges

not

man

aged

thr

ough

N

PA, s

o cu

ltura

l exp

ecta

tions

aro

und

retir

emen

t ag

es a

re n

ot m

anag

ed.

How

ever

, ind

exat

ion

adju

stm

ents

cou

ld

ince

ntiv

ise

mem

bers

to

mak

e th

eir

own

choi

ces

arou

nd r

etire

men

t ag

e.

Inde

xatio

n is

a c

ompl

ex a

rea

and

mem

bers

may

find

cha

nges

in

inde

xatio

n, a

nd h

ow t

hose

cha

nges

ar

e ca

lcul

ated

, diffi

cult

to u

nder

stan

d.

Link

a m

embe

r’s

NPA

to

thei

r cu

rren

t SP

A

timet

able

Man

ages

mos

t of

the

pas

t an

d pr

ojec

ted

cost

s as

soci

ated

with

ob

serv

ed a

nd p

roje

cted

incr

ease

d in

lif

e ex

pect

ancy

. Les

s ab

le t

o m

anag

e th

e un

cert

ainl

y ar

ound

long

evity

.

Due

to

the

unce

rtai

nty

arou

nd lo

ngev

ity t

here

is

unc

erta

inty

aro

und

whe

ther

the

cur

rent

SPA

tim

etab

le w

ill k

eep

up w

ith c

hang

es in

life

ex

pect

ancy

. If

it do

es n

ot t

he a

ssoc

iate

d co

sts

will

ne

ed t

o be

man

aged

thr

ough

oth

er m

easu

res.

The

defa

ult

wou

ld b

e fo

r sc

hem

e m

embe

rs

to k

eep

wor

king

unt

il SP

A; w

ould

ass

ist

in

crea

ting

a cu

ltura

l exp

ecta

tion

of c

hang

es

in le

ngth

of

wor

king

life

in r

espo

nse

to

chan

ges

in lo

ngev

ity.

Tim

etab

led

incr

ease

s to

SPA

intr

oduc

e so

me

com

plex

ity f

or m

embe

rs b

ut s

till

fairl

y ea

sy t

o un

ders

tand

.

Link

a m

embe

r’s

NPA

to

thei

r SP

A w

ith r

egul

ar

revi

ew o

f th

e lin

k

Man

ages

mos

t of

the

pas

t an

d pr

ojec

ted

cost

s as

soci

ated

with

ob

serv

ed a

nd w

proj

ecte

d in

crea

sed

in li

fe e

xpec

tanc

y. R

egul

ar r

evie

w

intr

oduc

es a

mec

hani

sm f

or d

ealin

g w

ith u

ncer

tain

ty s

urro

undi

ng lo

ngev

ity

Cos

ts a

ssoc

iate

d w

ith c

hang

es in

long

evity

mor

e lik

ely

to b

e m

anag

ed v

ia c

hang

es t

o th

e N

PA a

s ne

cess

ary.

Reg

ular

rev

iew

s m

ake

it m

ore

likel

y th

at

NPA

is s

et a

t a

fair

leve

l for

diff

eren

t co

hort

s.

The

defa

ult

wou

ld b

e fo

r sc

hem

e m

embe

rs

to k

eep

wor

king

unt

il N

PA, s

houl

d as

sist

in

crea

ting

a cu

ltura

l exp

ecta

tion

of c

hang

es

in le

ngth

of

wor

king

life

in r

espo

nse

to

chan

ges

in lo

ngev

ity.

Tim

etab

led

incr

ease

s to

SPA

and

th

e po

ssib

ility

of

futu

re c

hang

es

to N

PA in

res

pons

e to

cha

ngin

g lo

ngev

ity p

roje

ctio

ns in

trod

uce

som

e co

mpl

exity

. But

as

long

as

mem

bers

w

ere

give

n a

suffi

cien

t le

ad in

tim

e to

an

y fu

rthe

r ch

ange

it s

houl

d st

ill b

e fa

irly

easy

to

unde

rsta

nd.

NPA

au

tom

atic

ally

ad

just

ed in

line

w

ith c

hang

es in

se

t pa

ram

eter

s

Man

ages

mos

t of

the

pas

t an

d pr

ojec

ted

cost

s as

soci

ated

with

ob

serv

ed a

nd p

roje

cted

incr

ease

d in

lif

e ex

pect

ancy

. Aut

omat

ic a

djus

tmen

t w

ould

intr

oduc

e a

mec

hani

sm f

or

deal

ing

with

unc

erta

inty

sur

roun

ding

lo

ngev

ity.

Cos

ts a

ssoc

iate

d w

ith c

hang

es in

long

evity

sho

uld

be m

anag

ed v

ia t

he a

djus

tmen

t m

echa

nism

w

hich

sho

uld

also

ens

ure

that

NPA

is s

et a

t a

fair

leve

l for

diff

eren

t co

hort

s. B

ut o

ther

var

iatio

ns,

for

exam

ple,

Hea

lthy

Life

Exp

ecta

ncy

or s

ocio

-ec

onom

ic d

iffer

ence

s, n

ot a

ccou

nted

for

: cou

ld b

e pe

rcei

ved

as u

nfai

r.

The

defa

ult

wou

ld b

e fo

r sc

hem

e m

embe

rs t

o ke

ep w

orki

ng u

ntil

thei

r N

PA;

com

mun

icat

ion

over

cha

ngin

g N

PA s

houl

d as

sist

in c

reat

ing

a cu

ltura

l exp

ecta

tion

of c

hang

es in

leng

th o

f w

orki

ng li

fe in

re

spon

se t

o ch

ange

s in

long

evity

.

Diffi

cultl

y of

set

ting

the

auto

mat

ic

adju

stm

ent

fact

or: w

hat

long

evity

m

easu

re t

o us

e an

d w

hat

outc

ome

to a

im f

or, c

ould

be

perc

eive

d as

op

aque

. Cou

ld b

e co

mpl

ex f

or

mem

bers

to

unde

rsta

nd w

hat

the

adju

stm

ent

fact

or w

as b

ased

on

and

how

it w

orke

d.

Key

:

Not

met

Par

tially

met

Met

Sour

ce: I

PSPC

ana

lysis.

94


Recommendation: The Government should increase the member’s Normal Pension Age (NPA) in most schemes so that it is in line with their State Pension Age (SPA). However, the link between the SPA and NPA should be regularly reviewed to make sure it is still appropriate, with a preference for keeping the two pension ages linked (Recommendation 11).

4.13 Future pension earned by members would then have a NPA in line with their SPA. For example, based on the revised SPA timetable illustrated in Chart 4.C, a member aged 45 in 2011 would earn pension with an NPA of 66 and a member aged 25 in 2011 would earn pension with an NPA of 68. If the SPA timetable changed further in the future, then this would also change NPA for all pension earned under the new schemes.

4.14 Chart 4.D shows what these changes may mean for the expected proportion of life spent in retirement for public service employees if they retired according to the revised SPA timetable. The calculations are based on life expectancies used by HM Treasury in their long-term public finance report 2009, specific to public service employees, together with high and low variants based on ONS population projections.

Chart 4.D: Expected percentage of adult life in retirement after changes

20%

25%

30%

35%

40%

45%

1945 (66)

1950 (61)

1955 (56)

1960 (51)

1965 (46)

1970 (41)

1975 (36)

1980 (31)

1985 (26)

1990 (21)P

erce

ntag

e of

adu

lt li

fe s

pent

in

reti

rem

ent

Year of birth (age in 2011)

High life expectancy variant Central life expectancy projections

Low life expectancy variant

Source: IPSPC analysis, using life expectancies prepared by GAD for the National Audit Office.a

Note: Based on male SPA, under the revised timetable for increases – the female SPA is lower for those born before 1954.

Life expectancies are averaged between males and females.

a) National Audit Office: The cost of public service pensions. March 2010.

95


4.15 The chart illustrates how increasing NPA in line with SPA manages much of the increase in life expectancy we have seen in recent decades, and the improvements that are currently anticipated in the future. The central line representing HM Treasury’s central life expectancy assumptions for public service employees maintains the average expected proportion of life spent in retirement at around 33 to 35 per cent, broadly in line with the level it was in the 1980s (based on the NHS analysis above).

4.16 Of course, life expectancy is uncertain. Therefore, as well as the central estimates, the chart illustrates alternative scenarios using the higher and lower life expectancy assumptions the Government Actuary’s Department also prepared for HM Treasury (in line with high and low life expectancy projections from the Office for National Statistics). The chart shows that if life expectancy followed the high variant assumptions, the percentage could increase up towards 40 per cent, and if it followed the low variant assumptions, the percentage would be around 30 per cent.

4.17 As can be seen from the central life expectancy projection in Chart 4.D, once the final planned increase in SPA takes place in 2046 the expected proportion of adult life in retirement for public service workers is expected to start rising again. Without further action this would lead to increased costs due to life expectancy increases once again starting to rise over and above the allowance made in the proposed SPA timetable.

4.18 The uncertainty in future life expectancy, and the increase in costs after the last planned transition of SPA to 68 are risks that would ideally be managed in a system with NPA linked to SPA.

4.19 The Government has stated in its White Paper1 that “to manage the ongoing challenges posed by changes in projected longevity, the Government will be considering the current timetable for these rises and will bring forward proposals in due course.” Given this statement from the Government, the Commission considers that a link to SPA for the NPA of the public service schemes should help create a sustainable pension system that manages longevity risk.

4.20 However, the Commission’s recommendation is that as well as the link to SPA being put in place, NPA should also be regularly reviewed by an independent body, to see if the link is appropriately tracking changes in longevity. The body would then make recommendations to the Government (either for each scheme or for the public service as a whole) on whether linking the NPA for public service pension schemes to the SPA was still appropriate, and if not, what the NPA should be.

4.21 It is of great importance that any changes to the planned NPA timetable are fixed and publicised so as to allow a sufficient notice period for members. This would allow members to accommodate the change within their retirement planning and would give members more certainty and increase trust in the scheme.

1 DWP (2010) A sustainable State Pension: when the State Pension will increase to 66, Command Paper.

96


4.22 Members would also be given choice around their retirement age, as discussed in Chapter 3. The aim would be to move away from a ‘cliff edge’ retirement age as working practices and expectations around working life change. Whilst members could take an unreduced pension at their NPA there would be nothing to prevent them retiring earlier and taking an actuarially reduced pension, or indeed continuing to work past NPA, perhaps on a part-time basis, and continuing to accrue pension rights.

4.23 This approach should help to keep NPA within the public service pension schemes generally in line with developments in longevity and therefore ensure sustainability. A gradually changing NPA for the public service pension schemes broadly in line with changes in projected life expectancy would also assist in developing cultural expectations that as projected life expectancy changes so too should expectations of when to retire if people wish to maintain a steady standard of living in retirement.

Additional measures for controlling scheme costs

The need for additional measures

4.24 It has been the aim of the Commission to design a public service pension scheme structure which will share risks and costs between employees and government fairly.

4.25 The design features discussed in Chapter 3 and those regarding life expectancy in the sections above should achieve much of this aim. For example, moving to CARE from final salary removes much of the salary risk associated with final salary pensions. Adjusting NPA in line with longevity increases through linking to SPA will remove much of the risk to costs of future increases in longevity.

4.26 However, the Commission believes consideration should also be given to an overriding mechanism to ensure that public service pensions remain affordable and sustainable. This mechanism would act as a safety valve in case costs within the new scheme increased due to factors not taken account of in the scheme design.

4.27 This mechanism could be expressed as a ‘fixed cost ceiling’ and would be the upper limit on the amount that the Government would commit to employees’ pensions over the long term to each scheme.

4.28 The four largest public service pension schemes2 introduced cost control mechanisms as part of the 2007-08 reforms through cap and share arrangements.3 These arrangements were agreed between employers and trade unions and the intention was that certain increases in pension costs were shared between employer and employee up to a cap on employer costs. Introducing a cost ceiling would establish processes similar in some ways to cap and share

2 Teachers, NHS, Principal Civil Service and Local Government Pension Schemes.

3 See interim report pages 45-48 for a full description.

97


but which reflect the future scheme structure and have an automatic default change that will take place if agreement is not reached.

Recommendation: The Government, on behalf of the taxpayer, should set out a fixed cost ceiling: the proportion of pensionable pay that they will contribute, on average, to employees’ pensions over the long-term. If this is exceeded then there should be a consultation process to bring costs back within the ceiling, with an automatic default change if agreement cannot be reached (Recommendation 12).

4.29 The details of the control mechanism should be subject to an explicit consultation which should include:

• the process, timescales, parties involved;

• the level at which the ceiling is set;

• which scheme design features should be covered and what the automatic default change should be; and

• which cost factors are covered, in particular the extent to which past service costs are allowed for.

Defining a fixed cost ceiling

4.30 There are many ways of defining, measuring and setting a cost ceiling. However, the most appealing options for consideration are:

• pension expenditure each year as a proportion of GDP – essentially a cash measure; or

• the estimated cost of pension promises accruing over the next year as a proportion of total or pensionable pay bill – essentially an accruals measure.

4.31 One advantage of using the GDP measure is that in theory it will give a good degree of predictability for government about costs in the future and, since tax revenues are highly correlated with levels of GDP, it will also ensure that commitments remain affordable.

4.32 However, this measure is very difficult to control as it is influenced by a large number of variables outside the control of pension schemes costs such as future levels of GDP and the size of the public sector workforce. Having considered these factors the Commission has determined that this option is not viable.

98


4.33 The second measure is to look at the cost of the pension benefits accruing and keep this stable, either as a proportion of the total pay bill (i.e. for all employees whether in a pension scheme or not) or pensionable pay bill (i.e. for all pension scheme members).

4.34 Maintaining the pension cost under a ceiling based on total pay bill depends on the proportion of members choosing to participate in the scheme. Assessing pension costs as a proportion of pensionable pay bill avoids this impact of participation.

4.35 It is the Commission’s view that the proportion of pensionable pay approach is the most appropriate for the purpose of a cost ceiling.

Changing a scheme to remain under a cost ceiling

4.36 Increases in cost due to rising longevity should be managed through the linking of NPA to SPA. However, in the event that this mechanism is not sufficient or other factors lead to an increase in costs, then the main options for parameters which can be changed in principle are:

• an increase in employee contributions;

• a reduction in the accrual rate (for future service accrual only);

• a reduction in the indexation rate; and

• a reduction in the value of ancillary benefits (future service only).

4.37 These options are listed in decreasing order of simplicity to implement. Adjusting the accrual rate or the contribution rate for a particular year determines the amount of pension earned, or the contribution paid, in that year only. Conversely, adjusting the rate of indexation or the ancillary benefits is likely to have wider implications. For example, adjusting the indexation rate may affect all benefits that have been built up to date. Ancillary benefits are likely to be the most complex to adjust and require the most significant changes to yield the necessary savings.

4.38 The indexation option arguably has the advantage that it could potentially be applied to deferred and pensioner members rather than just active members, which could spread change more fairly, but adjusting pensions in payments has risks as pensioners may find it harder to manage the change. On balance, the Commission does not view varying indexation as an attractive option.

4.39 Considering the first two options further against the principles, there are conflicting arguments from an adequacy perspective. On one hand, it might be preferable to exclude the possibility of large increases in employee contributions which could lead to widespread opt-out amongst scheme members, particularly the lower paid and younger members. Conversely, it could be argued that adequacy is more affected by reducing the accrual rate.

99


4.40 This highlights a further difference in terms of cash flow. The contributions approach will have an impact on short-term cash flows only. The accrual rate approach would reduce the level of employer contributions in line with the constraints imposed by the ceiling. There would then also be a delayed impact as the pension to be paid in the future is reduced. Therefore, in terms of simplicity, employee contributions may be the more preferable approach as it only affects what is paid in, not also what is ultimately paid out.

4.41 Overall, the Commission considers that mechanisms which vary employee contributions or the accrual rate may be more preferable than looking to vary indexation or ancillary benefits.

Implementing a cost ceiling

4.42 A decision would also be needed on the level at which the costs are fixed: across all public service pension schemes or on an individual scheme by scheme basis. If set for all schemes then there would almost certainly be significant cross-subsidy between the different schemes. If costs rise in one scheme for whatever reason, benefits may need to fall or contributions to increase across all the schemes to compensate for this.

4.43 Therefore, it appears preferable that the cost ceiling mechanism should be set up and implemented at a scheme specific level, but that the process for operation should be common to all schemes. The process for the operation of the cost ceiling would need to be decided.

4.44 In the event of the evaluated cost of benefits exceeding the ceiling, the Commission’s view is that a default change should be generated. Government, employees and their representatives would then have the opportunity to discuss and negotiate a change that would meet the criteria of the ceiling. If no agreement could be reached then the default change would be implemented.

4.45 A key consideration is how often costs should be reviewed so that they remain under the ceiling. The initial choices may be:

• at each full actuarial valuation, carried out every three to four years. This is used to set the level of contributions required from employers and employees; or

• when the current service costs are calculated each year for the purposes of reporting in resource accounts. These are calculated using a corporate bond discount rate and so the published figures are subject to market volatility. However, there does not appear to be an obstacle to calculating an equivalent cost based on more stable assumptions such as that used in the SCAPE4 actuarial valuation process (following the ongoing review).

4 Superannuation Contributions Adjusted for Past Experience.

100


4.46 Whether a one-year turnaround time is practicable depends on the process set up to maintain the cost ceiling, for example, whether the system is automated to some extent or subject to negotiation.

4.47 The Commission’s view is that the cost ceiling process should run following each actuarial valuation, therefore every 3 or 4 years.

4.48 Valuations for most public service schemes are calculated using the SCAPE discount rate which has historically been 3.5 per cent above RPI inflation but is currently under review by the Government. The process of setting the parameters for a cost ceiling would need to consider the interaction between the eventual choice of discount rate and the calculated scheme costs.

4.49 For the Local Government Pension Scheme (LGPS), valuations are carried out separately for each fund with the choice of discount rate also set at the fund-specific level. Consistency of approach across the schemes would require the cost of pensions accruing within the different schemes to be calculated on a consistent set of assumptions (e.g. by applying the SCAPE discount rate to the LGPS valuations).

Dealing with past service

4.50 The design of the cost ceiling will need to identify the costs that should be shared with the public service membership rather than picked up by the taxpayer.

4.51 What types of cost should be included within the cost ceiling is a decision for the Government in consultation with employees and their representatives. The previous cap and share arrangements had broadly envisaged that increased costs in respect of demographic factors, such as salary increases and longevity, would be shared with members, while the Government would bear increased costs in respect of financial factors, such as discount rate, inflation and specifically to funded schemes, investment returns. This approach would be consistent with our analysis of where risks should sit within the public service pension schemes.

4.52 There are different approaches to which parts of pension costs should be included under a fixed cost ceiling approach, for example:

• the costs accruing as a result of service carried out in the year of cost evaluation (future service cost) only;

• the future service cost plus the past service costs relating to the currently active members of the new schemes;

• the future service cost plus the past service costs relating to all pension accrued to date within the new schemes; or

101


• the future service cost plus the past service costs relating to all pension accrued to date within the new schemes and the legacy (mainly final salary) schemes.

4.53 The first approach would only contain the costs accruing as a result of service carried out in future years. The three other options would cover increases in costs relating to past service. For example, if longevity increases, this will have costs associated for all years of service, which would not be picked up through only looking at future service costs. For past service within the proposed new scheme structure (options 2 and 3), this will be restricted to the longevity effects over and above the impact of adjusting NPA through the SPA link. To include the legacy schemes as well (option 4), the cost would cover the total longevity impact from a fixed NPA in the legacy schemes.

4.54 Past service cost might be difficult to estimate accurately in the unfunded schemes, but difficult to ignore in funded schemes since they will show up as part of the difference in the value of assets and liabilities. In the LGPS, the contributions towards the past service deficit are picked up by the employer.

4.55 There are significant pros and cons of including past service costs in the fixed cost ceiling.

4.56 Not including past service costs means that the Government (and ultimately the taxpayer) takes on all the risk associated with past service accrual. In fact, the only changes in cost to be shared would be those to come through from a change of assumptions. There would be no direct sharing of the cost impact of actual scheme experience unless the assumptions were also changed.

4.57 The general approach to past service within the cap and share agreements for the four schemes mentioned was that any changes to past service costs after the date of implementation of cap and share would fall within the cost share envelope. These agreements had been accepted by both the employers and the respective unions. Although it is not certain how these arrangements would have played out in practice, making no allowance for past service in the fixed cost ceiling would arguably be a weakening of the cost sharing agreements that had been agreed.

4.58 Excluding much of the past service costs is a view supported by the Railway Pension Commission who recommended that all costs associated with increased longevity of deferred and pensioner members be borne by the employers.5 If this approach was adopted it would mean this element of past service cost associated with changes in longevity would sit outside the cost ceiling and these costs would fall to the Government.

4.59 Including the past service costs within the cost ceiling would mean that the increased cost of accrued rights would need to be shared among the active scheme members, which will have negative implications for the benefit levels or contributions required from this group. For example, active members may not only bear the impact of increases in past costs

5 Railway Pensions Commission (2008), Final Report.

102


that they themselves may benefit from (which may be nothing, if a new member), but also any increases which deferred and pensioner members are benefiting from.

4.60 The greater the extent that past service is included within the fixed cost ceiling, the greater the volatility of employee contributions or benefit rates. The effects, particularly those relating to pensioner and deferred members, are emphasised in mature schemes (those with a high number of pensioner and deferred members in relation to active members). This is because the increased cost of accrued rights will need to be shared amongst a relatively small active population.

4.61 The changes may discourage participation within the schemes. There may be an argument for targeting contribution increases at those who do have past service, but again there is nothing to stop these members leaving the scheme.

4.62 If it was felt that future service costs and past service costs could not be dealt with within the same cost ceiling, it may be worth considering a more tailored solution such as measuring increases in future service costs against one ceiling and capping the impact of past service costs at another. Given past discussions and agreements on this question, the Commission has concluded that the extent to which past service costs are allowed for in the fixed cost ceiling should be part of the consultation process following this report.

103


5 Applying the design

Box 5.A: Summary

There are differences between the individual public service schemes in factors such as the distribution of pensionable pay, average career length and life expectancy. But, for most schemes, these differences are generally similar to, or smaller than, those seen within the schemes. The modernised schemes generally apply standard features to their members and this is accepted as an appropriate approach, as tailoring schemes for different groups creates a complex and costly system and would be likely to reduce member understanding. This would suggest that, for most schemes, variation in pension features is not the most appropriate way to deal with these differences where they are seen.

The Commission is not proposing a single public service pension scheme, but over time public service pensions should move towards a common framework for scheme design as set out in this report. However, in some cases, for example, the uniformed services, there may need to be limited adaptations to this framework.

For the uniformed services, expectations and life expectancy, have moved on since the Normal Pension Ages (NPAs) of 55 or less that are seen for the majority of long-serving members of the uniformed services schemes, were set. However, there is a need to recognise the unique nature of the work the uniformed services (the armed forces, police and firefighters) undertake. Therefore, in the case of the uniformed services schemes the Government should consider setting a new NPA of 60, where the NPA is currently below this level, and keep this under regular review, rather than linking their NPA to State Pension Age (SPA).

The common framework should apply to the LGPS, where local needs should also be met mainly through varying pay and allowances. However, it remains appropriate for the Government to maintain the different financing arrangements for the LGPS in future, so the LGPS remains funded and the other major public service schemes remain unfunded. This would maintain a mixed approach to the funding of the public service pension schemes.

As for who in future should be entitled to join these schemes, it is ultimately for the Government to decide how much long-term pensions risk it is willing to bear in order to meet its wider policy objectives. However, it is in principle undesirable for future non-public service workers to have access to public service pension schemes, given the increased long-term risk this places on the Government and taxpayers.

104


5.1 The public services include a very wide range of employees. This diversity could potentially affect the precise application of the Commission’s proposed reforms, summarised in Box 5.B below, and the transition to new arrangements.

Box 5.B: Key design features in the common framework

Based upon the four key principles described in Chapter 1, the Commission believes that future public service pension schemes should have the following features:

• pensions should be defined benefit, based on career average revalued earnings;

• pensions in payment should be indexed by inflation, while accrued benefits should be revalued by earnings for active members;

• accrual rates, indexation methods and employee contribution levels should be considered together, to ensure that pension schemes deliver adequate retirement incomes and are sustainable;

• a member’s Normal Pension Age (NPA) for future accruals should be linked with their State Pension Age (SPA); and

• the Government should introduce a cost ceiling, to help control for future changes in longevity and other costs.

The extent of variation across schemes and across their membership5.2 The interim report1 noted that the development of public service pension schemes has not been a planned and fully coherent process and that there is a plethora of complex provisions. A wide range of professions are covered by different schemes but also within the same scheme. Again, as the interim report noted,2 different schemes, designs and contributions apply to people employed in similar public service jobs, sometimes for the same employer, for example, teachers are generally in the Teachers Pension Scheme (TPS)and teaching assistants in the Local Government Pension Scheme (LGPS).

5.3 And information technology (IT), accountancy and public administration are found as professions across the public services. Within the civil service scheme groups as diverse as lawyers, economists, scientists and engineers are covered alongside groups such as border officers, coastguards, prison officers and some criminal investigators and police support staff. The situation is similar for most of the other public service schemes.

1 IPSPC (2010) Interim report, paragraph Ex.2.

2 IPSPC (2010) Interim report, paragraph 1.10.

105


5.4 The great variation within schemes and the common features generally shared between schemes are an inevitable result of providing very large schemes covering much of a particular sector.

5.5 Schemes also display major common features that reflect centrally agreed pension policies, such as final salary design, multiple tiers of ill health pension, or flexibilities to take early pension.

5.6 There are differences in how approaches to pension design have been applied in practice, reflecting particular scheme characteristics. For jobs with a physical element, particularly in the uniformed services (the armed forces, police and firefighters), pension ages of 55 or less have been used to recognise the effects of ageing and limitations on longevity. Also, pension ages of 40 or less have been used by the armed forces as a device to aid retention and encourage exit. And the uniformed services in general have not adopted flexible retirement options, bearing in mind that unreduced immediate pensions have generally been available under their schemes’ rules by the age that flexible retirement options would have become available to them.

5.7 However, whilst such factors are still important they are not as significant as they once were. This is, for example, reflected in the increases in longevity seen across all groups and in some of the changes already made to public service pension terms. These include the pension age of 653 that applies to new entrants to professions such as nursing or custody and care of prisoners and the normal pension age of 60 in the Firefighters Pension Scheme 2006 for those who serve until 60. And early leavers in the uniformed services now generally have a NPA of 65.4

5.8 Differences between the professions in the extent of work-related risks tend to be reflected in differences in pay and allowances and in the compensation terms that are provided, so, for example, the uniformed services have more valuable death and injury cover.

5.9 Alongside that, there are major differences in career and remuneration structures between groups within the same scheme that are not reflected in pension design. Those differences can be as great as or greater than the differences in career and remuneration between members of different schemes.

5.10 The average annual earnings within the six biggest public service schemes (police, teachers, armed forces, National Health Service (NHS), civil service and the LGPS) range from around £20,000 to nearly £40,000.5 However, this variation in average earnings between the schemes conceals wide variations in earnings within the schemes and similarities between the schemes in their pay distribution. This is shown in Charts 5.A and 5.B below.

3 Instead of 55 or 60.

4 Covering those who leave the new firefighters scheme before age 60, those who leave the Police Pension Scheme 2006

before age 55 and those who leave the Armed Forces Pension Scheme 1975 before about age 40 or the Armed Forces

Pension Scheme 2005 before age 55.

5 Data returns to IPSPC except NHS: 2007 Scheme valuation and 2009-10 resource accounts; and civil service.

106


The armed forces, teaching and the police have fairly similar distributions (though with a significant spike of low earners in the armed forces). There are also significant similarities between the NHS, civil service and local government schemes, although there are comparatively more high earners in the NHS.

Chart 5.A: Distribution of pensionable pay in armed forces, teachers and police schemes

0%1%2%3%4%5%6%7%8%9%

10%

10 15 20 25 30 35 40 45 50 55 60 65 70 75 80

Per

cen

t of t

otal

act

ive

mem

bers

Full-time pensionable earnings (£000)

Armed forces Teachers Police


Note: Police data as of 31 March 2008. Armed forces rates include pay during basic training.

107


Chart 5.B: Distribution of pensionable pay in NHS, civil service and local government schemes

0%1%2%3%4%5%6%7%8%9%

10%

10 15 20 25 30 35 40 45 50 55 60 65 70 75 80

Per

cen

t of a

ctiv

e m

embe

rs

Full-time pensionable earnings (£000)

NHS Civil service LGPS


5.11 However, there are large differences in the average career lengths for scheme members, ranging from 7 to 8 years for the LGPS, around 10 years for the armed forces, 11 for the NHS and 13 for the civil service to about 16 for judges, 18 for firefighters, 23 for teachers and nearly 25 years for the police.6 These averages would represent from under a quarter to over a half of a potential working lifetime.7

5.12 But, again, those averages disguise large variations in career length between members within schemes. Full information is not available at present for many of the schemes, but those for which comprehensive data have been obtained are included in Chart 5.C below.

6 NHS - NHS staff turnover statistics 2007-08. Teachers - School workforce in England 2010. Civil service - tables prepared

for IPSPC by Office for National Statistics. LGPS - Local government employers sample survey 2007. Armed forces

- UK Regular Forces Intake and Outflow by Age 2009-10. Police - Home Office Statistical Bulletin, 22 July 2010. Fire -

Operational statistics bulletin for England 2009-10. Judiciary - Data return to IPSPC from Ministry of Justice.

The LGPS figures are based on all local government workers. Since relatively many workers choose to opt out of the LGPS

scheme, and those who opt out are likely to have shorter careers on average, the average career length of those who stay

in the scheme is likely to be somewhat longer. For comparison, the average career length of current local government

pensioners is 16 years (source: local government scheme data returns).

In most schemes average pensionable service is also greater than these figures for average career lengths because those

who serve for a short time do not meet the minimum qualifying period required to get a pension from the scheme: two

years for most schemes and three months for the LGPS.

7 Assumed here to mean the period between leaving full-time education and reaching State Pension Age.

108


Chart 5.C: Variation of career lengths in public service pension schemes

0

2

4

6

8

10

12

14

16

0 5 10 15 20 25 30 35 40

Perc

enta

ge o

f mem

bers

Years of service

Teaching LGPS Judiciary Armed forces Civil service

Source: Scheme data returns to IPSPC, except Armed Forces – provisional data provided by Service Personnel and Veterans Agency for 2009 valuation.

Notes: The LGPS figures are based on pensionable service. Given the high proportion of part-time employees in the LGPS,

this is likely to understate significantly actual years of service.

Armed forces figures reflect that there are many engagements that last either 4 years or around 20 years, when

Immediate Pensions have been paid.

Judiciary figures reflect that most judges retiring in the last two decades were entitled to pensions of half pay after 15

years’ service.

5.13 This chart indicates that there is a large spread in the length of service seen within the public service schemes, notwithstanding the peaks seen in some schemes, which are driven mainly by employment or pension terms.

5.14 Life expectancy assumptions and experience do vary somewhat between schemes. Chart 5.D shows the projected life expectancy at 60 for those schemes for which we have comparable data. However, the variations in life expectancy are generally not that marked and the life expectations are above the average for the population as a whole.

109


Chart 5.D: Projected life expectancy at age 60

0

5

10

15

20

25

30

35

NHS Teachers Civil service Armed forces (officers)

Armed forces (other ranks)

Yea

rs

Male life expectancy at age 60 Additional female life expectancy

General UK population (Males) General UK population (Females)

Source: Scheme life expectancies from resource accounts, 2010. General population life expectancies published by the ONS.

Notes: 2010 projections are not yet available for other schemes. Life expectancies are all from age 60, for someone

reaching age 60 in 2010. General population statistics are cohort expectations of life, based on historical mortality rates

from 1981 to 2008 and assumed calendar year mortality rates from the 2008-based principal projections.

5.15 Most schemes do not have scheme specific data showing the underlying differences in life expectancy between members, but these are likely to reflect the different life expectancies of the different categories of workers within the schemes. And, as discussed above, there is a large variation in the types of worker represented within the different schemes. The types of variation in life expectancy that could be expected based on job classification are outlined in Chart 5.E, and this variation is greater than that seen in the averages between schemes.

110


Chart 5.E: Male life expectancy at age 65 by social class

1011121314151617181920

1972-76 1977-81 1982-86 1987-91 1992-96 1997-2001 2002-2005

Years

Professional Managerial and technical

Skilled non-manual Skilled manual

Semi-skilled Unskilled

Source: Office for National Statistics longitudinal survey.

Standardisation and variation in core designs: one or many

schemes

5.16 The evidence presented above suggests that there are differences between the individual schemes in factors such as the distribution of pensionable pay, average career length and life expectancy. But, for most schemes, these differences are generally similar to, or smaller than, those seen within schemes in these parameters.

5.17 The modernised schemes generally apply standard features to their members, for example, having one NPA or accrual rate. This is accepted as an appropriate approach, as tailoring schemes to all the differences seen would create a complex and costly system and would be likely to reduce member understanding. This would seem to suggest variation in pension features is not the most appropriate way to deal with these differences where they are seen, for the majority of schemes.

Recommendation: The Commission is not proposing a single public service pension scheme, but over time public service pensions should move towards a common framework for scheme design as set out in this report. However, in some cases, for example, the uniformed services, there may need to be limited adaptations to this framework (Recommendation 13).

111


The different position of the uniformed services

5.18 Despite the wide variations in membership characteristics within most schemes, most have handled pension ages as standard features across the membership rather than tailoring them to different groups within the workforce. Schemes have therefore generally adopted a NPA of 60 or 65 across the whole membership.

5.19 The only significant exceptions to this have been the use of NPAs of 55 or less for the members of the uniformed services who serve until that age although, as discussed in paragraph 5.7, this has been changing for deferred members of the uniform services and for members of the modernised firefighters scheme.

5.20 However, the pension ages in the uniformed services schemes still generally reflect an assumption that pension for the majority of long-serving members should be payable from age 55 or less. But this assumption may no longer match expectations, given the increases in life expectancy that have been seen since the 19th and first half of the 20th century; when these pension ages were set.

5.21 It also seems that leavers from schemes such as the armed forces, police and firefighters are often well qualified for careers elsewhere and do not suffer from high levels of unemployment.

5.22 For example, a National Audit Office survey in 2007 found that only six per cent of all leavers from the armed forces were unemployed.8 Similarly, Ministry of Defence surveys of service leavers show that 94 per cent of those seeking work who used the Department’s resettlement service9 were employed within six months.10 Those most prone to unemployment were those who left after less than four years’ service or on compulsory discharge. In the case of firefighters, exit surveys show that more than two-thirds of those who resign before retirement do so in order to take up other employment.11

8 National Audit Office (2007), Leaving the services.9 Career Transition Partnership.

10 National Audit Office (2007), Leaving the services.11 Fire and Rescue Service (2010), Operational Statistics Bulletin for England 2009-10.

112


Chart 5.F: The employment status of armed forces leavers

Paid full-time workPaid part-time workEducationUnemployed, wanting workOther

Source: National Audit Office (2007), Leaving the services.

5.23 There is less evidence about the employment of those above current NPAs, but we do know that there are currently about 1,200 active police officers over the age of 55 working in UK police forces.12

5.24 However, this does not take away from the fact that the nature of the work the uniformed services perform is unique and that this needs to be reflected in their NPAs. But, as discussed above, things have changed since the NPAs for the majority of long-serving members were set. This has been recognised in the modernised firefighters scheme and the Commission’s view is that the NPA in this scheme, 60, should be seen as setting a benchmark for the uniformed services as a whole. This position will need to be kept under regular review to make sure it is still appropriate, given future changes in life expectancy projections and experience of healthy life expectancy.

Recommendation: The key design features contained in this report should apply to all public service pension schemes. The exception is in the case of the uniformed services where the NPA should be set to reflect the unique characteristics of the work involved. The Government should therefore consider setting a new NPA of 60 across the uniformed services, where the NPA is currently below this level in these schemes, and keep this under regular review (Recommendation 14).

5.25 However, where future deferred members of the current uniformed services schemes already have a NPA of 65 it seems appropriate to link their NPA to the SPA in the new schemes, as suggested for the majority of public service schemes in Chapter 4. As SPA might reasonably be viewed as when someone could be expected to end their working life.

12 Ministerial written answer, Hansard HC Deb, 9 November 2010, c200W.

113


Variations across the UK Government Administrations

5.26 Although pensions policy, including public service pensions policy, is set at a national level, a number of the public service pension schemes are the responsibility of the Devolved Administrations rather than the UK Government. There has been scope for some variations in terms to meet local circumstances, but the resulting pension schemes have essentially been the same as those established by the UK Government. That has, for example, helped to prevent pension terms becoming an obstacle to transfers of staff and skills within a sector of the public service. It seems reasonable to continue with this approach.

5.27 The key design features should be part of a UK-wide policy framework that extends to Scotland, Wales and Northern Ireland, with limited adaptations of other features to meet local circumstances.

Impact on retention

5.28 The Commission recognises that there will be a need to deal with specific recruitment and retention requirements that vary within and between workforce groups over time and that cannot, and should not, be dealt with through the common design framework recommended in this report.

5.29 The Commission’s view is that major variations in job demands and management needs for particular recruitment and retention effects should usually be reflected in pay and allowances, including lump sums payable after some years’ service, rather than special pension terms. Many such retention and reward incentives already exist and are usually more flexible than pensions. They can also be tailored more easily and quickly to reflect necessary skills, changes in the ages and backgrounds of new entrants and any special, often temporary, demands. However, because offering up-front pay and lump sums instead of higher pensions in future can bring spending forward, it would be necessary to manage any change carefully.

Funding and local government pensions

5.30 The interim report13 recognised that the Local Government Pension Scheme (LGPS) is in a different position to other very large public service schemes because it is funded, so contributions for employers and employees are invested and pensions are paid out of the 101 different LGPS Funds across the UK.

5.31 That requires some differences in how the scheme is managed. However, as the interim report also noted,14 the LGPS provides a similar set of final salary-based benefits to unfunded

13 IPSPC (2010) Interim Report, paragraph 4.76.

14 IPSPC (2010) Interim Report, paragraph 4.94.

114


schemes. There are overlaps in coverage with unfunded schemes, which sometimes cover similar workforces or other employees of an employer who also employs LGPS members.

5.32 The assessment of variations between and within schemes in this report seems to bear out that the LGPS membership shares many features with other schemes. For example, the members on average or higher earnings are comparable to those in other schemes and although the LGPS has a higher proportion of part-time, lower earning members, there are also many such members in schemes, for example, the civil service and NHS schemes.

Recommendation: The common design features laid out in this report should also apply to the LGPS (Recommendation 15a).

5.33 The core benefits package would therefore apply to the LGPS as to unfunded schemes, with variation in the remuneration package mostly being handled through pay and allowances. 15

5.34 On the issue of whether schemes should be funded or unfunded, the interim report suggested that it was reasonable, on balance, for government to structure public service pension provision on a mainly unfunded basis.16 Funding is not needed to safeguard statutory benefit promises backed by the taxpayer and effective, transparent long-term fiscal planning can be used to help manage risks. This is discussed further in Chapter 6. It is doubtful whether increasing government taxation or explicit borrowing to invest monies in pension funds would lead to higher economic growth. Funding schemes through gilt issuance would probably result in an increase in the cost of government borrowing and there would be investment management costs and risks if the schemes moved to a funded basis. And, as the interim report noted,17 the transitional cash flow effect of moving to funded pensions could be £20 billion or more a year for many years.

5.35 However, the interim report also suggested that the current LGPS should continue on a funded basis.18 One issue that the Commission subsequently considered was that whether it might be possible to consider moving to an unfunded basis for future LGPS service at the point that the reform package recommended in this report was implemented. That would then mean that, for future service, the LGPS might be managed like the unfunded police and firefighters schemes that are also administered by local government.

15 Most of the processes for assessing future and past costs and long-term sustainability and employee contributions

would also be comparable to those in unfunded schemes. For example, there would need to be a transparent

consolidation of valuation information across the LGPS, key assumptions that were provided and used on a consistent

basis across schemes, assuming that investment risk were to be excluded from the LGPS fixed cost envelopes, and

negotiation and consultation on implementing the results. Such processes were underway in planning for cap and share,

as discussed in Chapter 2 of the interim report.




115


5.36 That approach could provide some additional cash for central government in the short to medium term and would eventually remove the annual fund investment management costs of around £300 million. However, it would also increase the deficits across the LGPS funds (as they would no longer get the employer and employee contributions) and the resulting pressures would need to be made good by government at some stage. Also around a quarter of LGPS members are not employees of local authorities;19 this approach would place these members into an unfunded scheme.

Recommendation: It remains appropriate for the Government to maintain the different financing arrangements for the LGPS in future, so the LGPS remains funded and the other major schemes remain unfunded (Recommendation 15b).

Access to public service pension schemes

5.37 In recent years, there has been a drive to encourage alternative models of public service delivery. This has led to services that were traditionally delivered by the public sector being increasingly delivered by outside bodies. This trend is likely to continue, as the current Government is interested in pursuing opportunities for increasing the involvement of the private sector, voluntary sector and mutual organisations in the delivery of public services.

5.38 The Commission, in its interim report,20 raised some of the issues related to this plurality of public service provision, particularly in relation to the Fair Deal policy. Since then, the Government has established reviews of Fair Deal and the discount rate, which will examine the issues we raised in more detail. In this section we will review current scheme membership arrangements and their implications.

Overview of current scheme membership details

5.39 The Commission has worked closely with the largest public service pension schemes to understand the landscape of their membership. Whilst this has produced useful and informative data, there have been gaps in the availability of data and notable differences in the amount of data available between the schemes. This has complicated comparisons between the schemes and has made it harder to achieve a full understanding of the membership landscape. In future, the Commission believes there is a need for schemes’ membership information to be systematically collected and readily accessible, given the potential implications for government.

5.40 Table 5.A below sets out the UK-wide headline membership information we have received across the local government, NHS, and teachers schemes. We have concentrated on

19 They are mostly private sector employees but some are employed by other public sector employers.


116


these because they are the largest public service pension schemes with significant non-public service membership.

Table 5.A: Headline membership information for some of the largest public service pension schemesa

Scheme Total number of employers

Number of admitted organisations

Membersin whole scheme (million)b

Members in an admitted body (%)

Total active membership for whole scheme (%)

Average membership per employer

LGPS 7,942 6,3832 4.8 23.4c 41 609

NHS 10,506 453 2.7 1.9 55 258

TPSd 2,870 1,421e 1.7 6.3f 34 608


a Total and average membership figures include deferred and pensioner members.

b Total non-local government employers, including scheduled bodies.

c Total non-local government employers, including scheduled bodies.

d Information relates to teachers’ last period of service. Some teachers will have moved between sectors during their careers

(particularly the independent and local authority sectors) and some deferred TPS members and pensioners may have worked

in independent schools that are no longer within the scheme.

e Independent schools.

f Independent schools.

5.41 Of all public service pension schemes, the LGPS has the largest membership at 4.8 million and the highest number of additional organisations, with more than 6,000 non-local authority bodies in the scheme. This includes organisations from the private and voluntary sectors, as well as a range of public sector bodies such as housing associations, fire authorities, police authorities, colleges and universities. Almost 25 per cent of LGPS members belong to bodies other than local authorities; these organisations are generally smaller than local authorities, with an average membership (including deferred and pensioner members) of 177, compared to 2,375 in local authorities (a figure that is in itself reduced by the inclusion of many small town and parish councils). Active membership in these bodies is 45 per cent, slightly higher than the overall active membership of 41 per cent for the whole scheme. The LGPS is the only scheme that operates a shorter vesting period21 of three months (most public service pension schemes have a two-year vesting period), which increases the proportion of deferred members compared with active members.

5.42 The NHS scheme has a much smaller proportion of admitted organisations (Direction Bodies), comprising 4 per cent of employers and 2 per cent of members. The scheme also includes over 9,500 self-employed General Practitioner (GP) practices, with an average of

20 If members leave the employment during the vesting period they will receive a contribution refund or a transfer value

but they do not have the right to become a deferred member.

117


8 members in the NHS scheme. Direction Bodies themselves have an average membership of 116. Overall, the NHS scheme has the highest proportion of current active members amongst the three public service pension schemes in Table 5.A, at 55 per cent. The level of active membership within Direction Bodies (54 per cent) is in line with the scheme’s overall average.

5.43 In contrast the TPS has the lowest overall proportion of current active members, at 34 per cent. There is a much higher proportion of active membership in Scotland (51 per cent) and Northern Ireland (47 per cent) than in England & Wales (33 per cent). There are more than 1,400 independent schools in the scheme, with an average of 77 members (including deferred and pensioner members). The level of active membership within independent schools (44 per cent) is 10 per cent higher than the level of active membership for the scheme as a whole.

5.44 It is worth noting that there is likely to be some double counting in these numbers. For example, there may be some double counting in employer numbers because employees of the same organisation can be members of different schemes; teachers in local authority schools are members of the TPS but teaching assistants and support staff are members of the LGPS, to give just one example.

5.45 The Commission believes that the landscape presented here is the most comprehensive produced to date but more would need to be done to understand the full membership picture of public service pension schemes. In the future, it would be helpful for government to have a complete breakdown of the membership of these schemes, in particular in relation to the precise level of membership from employees in the private sector.

Overview of current scheme membership arrangements

5.46 Some public service pension schemes have long histories of extending their access to non-public service employees, but this extension of access has not happened consistently across the schemes.

5.47 As noted earlier, the LGPS has the highest number of additional organisations of any public service pension scheme. There are more than 6,000 such bodies in the LGPS scheme, covering 23.4 per cent of LGPS members. These include contractors that take on local authority services (transferee admission bodies), charities and non-profit organisations (community admission bodies) and a range of other public sector organisations. Admitted bodies may be required to provide the scheme with an indemnity or bond, if they are considered to be at heightened risk of defaulting on their pension commitments. They are also likely to be charged a higher contribution rate than a local authority (because their pension liabilities are typically funded over the shorter contract period) and the participation terms (including any risk-sharing mechanism) will depend on the terms of their admission agreement.

118


5.48 The teachers scheme also includes a wide range of organisations, almost 3,000 across the UK, including local authorities, higher and further education establishments, academies and independent schools.22 There are more than 1,400 independent schools, covering just over 6 per cent of the membership. Independent schools do not receive funding from government. It is believed that the majority of independent schools are within the private sector and also hold charitable status. Independent schools are required to provide evidence of a financial guarantee, indemnity or bond to be accepted into the scheme. All teachers employed by academies automatically become members of the TPS, unless they have elected to opt out of the scheme. Non-teaching staff (including teaching assistants) employed by local authority schools are members of the LGPS, but other teaching employers will often make different pension arrangements for their non-teaching staff.

5.49 The NHS scheme includes the staff of Direction Bodies,23 which are additional organisations approved to join the scheme by the Secretary of State. Under current policy, Direction Bodies are not allowed to be profit-making and in the majority of cases non-NHS work carried out by staff cannot be pensioned. The NHS scheme also includes self-employed GPs and their staff, as well as Dental Practitioners. As noted earlier, Direction Bodies make up a smaller proportion of additional organisations than the teachers and local government schemes. However, it is also worth noting that 91 per cent of scheme employers are self-employed GP practices.

Implications of access rules on public finances

5.50 The membership of the major public service pension schemes has expanded over time, responding pragmatically to changing circumstances. This has resulted in a variety of people and organisations entering the schemes that was not always envisaged when those schemes were set up, including a significant number of private sector employers and staff.

5.51 This expansion has happened inconsistently. Different professions within the same organisation can be members of different schemes and different organisations within the same sector can be members of different schemes.

5.52 Increasing access to public service pension schemes has had practical benefits, in terms of enabling non-public sector organisations to take over public sector functions, services and transferred staff. As a result, there are now many public service pension scheme members who do not work within the public sector.

5.53 There are clear pros and cons to allowing access to public service pension schemes. In terms of advantages, enabling access helps to remove the pensions barrier for external contractors with in-house services. This can enable more transparent contract prices, as bids can focus on the costs of service delivery rather than pension provision. Enabling access

22 Registered under Section 161 of the 2002 Education Act.

23 The direction is made on a case-by-case basis by the Secretary of State, under Section 7 of the Superannuation

(Miscellaneous Provisions) Act 1967.

119


can facilitate the transfer of staff to new employers and can also maintain the cash flow of contributions into schemes, which is particularly important for the unfunded schemes.

5.54 However, there are also clear disadvantages to enabling access, as it increases the Government’s risk of taking on liabilities for a workforce that it does not control. By allowing external organisations into public service pension schemes, the Government is at risk from the financial consequences of those organisations failing, awarding excessive pay rises, being unable to pay their exit fees when they leave schemes or of them paying insufficient contributions whilst they are in the scheme for the long-term liabilities that their members accrue.24

5.55 Given that the Government is keen to extend alternative models of public service delivery, the issues related to public service pension scheme access are likely to be of relevance to an increasing number of people and organisations in the years ahead.

5.56 As mentioned above, since the publication of the Commission’s interim report the Government has announced reviews of the Fair Deal policy and the discount rate, which are relevant to many of the issues discussed here. The Commission expects that the outcome of these reviews would, at least in part, help to facilitate the Government’s aim for increased plurality of provision for public services.

5.57 A redefined public service pension scheme framework, as laid out in this report, including a move to schemes based on career average revalued earnings, should over time also help to remove some of the barriers to plurality of service provision.

5.58 It is ultimately for the Government to decide how much long-term pensions risk it is willing to bear in order to meet its wider policy objectives. However, it is clear that enabling access to public service pension schemes for non-public service workers does increase the long-term risk government bears in relation to those schemes.

Recommendation: It is in principle undesirable for future non-public service workers to have access to public service pension schemes, given the increased long-term risk this places on the Government and taxpayers (Recommendation 16).

5.59 The issues concerning access to public service pension schemes are complex and wide-ranging. Enabling access to public service pension schemes has clear pros and cons and it will ultimately be for the Government to consider how best to address these issues, in the light of its wider policy priorities.

24 Although it should be noted that some schemes have rules in place to mitigate these risks.

120


121


6 A transparent and effective system

Box 6.A: Summary

• The publication of public service pension scheme data is inconsistent and this hinders effective monitoring and analysis. All public service pension schemes should regularly publish data which, as far as possible, is produced to common standards and methodologies and is then collated centrally. This information should be of a quality that allows simple comparisons to be made across government, between schemes and between individual Local Government Pension Scheme (LGPS) Funds.

• Governance arrangements for public service pension schemes vary considerably. All scheme members deserve to know that their scheme is being properly run and every public service pension scheme (and individual LGPS Fund) should have a properly constituted, trained and competent Pension Board, with member nominees, responsible for meeting good standards of governance, including effective and efficient administration. There should also be a pension policy group for each scheme at national level, for considering major changes to scheme rules.

• Communication with scheme members is considered crucial in improving general pension knowledge and in promoting a sense of pension ‘ownership’. All public service pension schemes should issue regular benefit statements to active scheme members, at least annually and without being requested and promote the use of information technology (IT)for providing information to members and employers.

• Governance and the availability and transparency of information would be improved by government establishing a framework that ensures independent oversight of the governance, administration and data transparency of public service pension schemes. Government should consider which body or bodies, including for example, The Pensions Regulator, is most suitable to perform this role.

• When assessing the long term sustainability of public finances it is important that the impact of public service pensions is subjected to closer scrutiny than is currently the case. The Office for Budget Responsibility should provide a regular published analysis of the long term fiscal impact of the main public service pension schemes (including the funded LGPS).

• Managing investment funds is an additional aspect of pension scheme governance for the funded LGPS schemes. Currently the funding and investment strategies are too narrowly focused on the individual Funds rather than on the overall sustainability of the LGPS. Centrally collated comprehensive data, covering all LGPS Funds, should be published including Fund comparisons, which, for example, clarify and compare key assumptions about investment growth and differences in deficit recovery plans.

122


Box 6.A (continued): Summary

• Good administration is essential if pension schemes are to be run well. Many schemes seem to be, but costs vary significantly and meaningful comparisons are hampered by a lack of defined standard outputs. Government should set what good standards of administration should consist of in the public service pension schemes based on independent expert advice. The Pensions Regulator might have a role, building on its objective to promote good administration. A benchmarking exercise should then be conducted across all the schemes to help raise standards where appropriate.

• New initiatives to save costs by sharing administrative services and contracts are being trialled by a number of LGPS authorities across the UK. Central and local government should closely monitor the benefits associated with the current co-operative projects within the LGPS, with a view to encouraging the extension of this approach, if appropriate, across all local authorities. Government should also examine closely the potential for the unfunded public service schemes to realise greater efficiencies in the administration of pensions by sharing contracts and combining support services, including considering outsourcing.

• The current public service pension schemes are established under a variety of legislative arrangements. In order to provide greater transparency, simplicity and certainty, the Government should introduce primary legislation to adopt a new common UK legal framework for public service schemes.

The need for change

6.1 In its interim report1 the Commission noted that the debate around public service pensions is hampered by a lack of consensus on key facts and figures and a lack of readily available and relevant data. There are also inconsistent standards of governance across the schemes. Consequently it is difficult for scheme members, taxpayers and commentators to be confident that schemes are being effectively and efficiently run. It also makes it more difficult to compare between and within schemes and to identify and apply best practice for managing and improving schemes.

6.2 This chapter looks at the current arrangements for the overall governance, administration and financial management of public service pension schemes, including the publication of data, comparing these where appropriate with the arrangements for private sector occupational pension schemes. It recommends improvements which the Commission believes will help improve both trust and confidence in the way the schemes are managed.

6.3 The legal and regulatory framework that applies in the private sector would not be appropriate for the public service given, for example, the statutory roles of Secretaries of State and others and the effective underwriting of unfunded and funded public service


123


pension promises by the State. However, there seems no good reason why the main principles and some of the best practice of the private sector should not also be applied in the public services. The Government might also consider whether such changes should be framed by new overarching primary legislation, which would replace provisions such as the Superannuation Act 1972.

Governance

Current governance arrangements for public service pension

schemes

6.4 Currently the governance arrangements for public service pension schemes vary considerably with some schemes having structures in place that are as good as benchmarked examples in the private sector, whilst others are not yet to this standard. There are various categories of governance structures at present, which can broadly be categorised as unfunded and centrally administered, unfunded and locally administered and funded and locally administered. These are described in Table 6.A.

Table 6.A: Types of governance structures for public service pension schemes

Unfunded centrally administered e.g. teachers, NHS, armed forces

Unfunded locally administered i.e. police & firefighters

Funded locally administered i.e. LGPS

The appropriate Secretary of State or Minister is legally responsible for the scheme and sets policy and rules, but delegates the day to day running to an Accounting Officer, accountable to Parliament.a Some schemes have a pension board or group.

The relevant Secretary of State (Home Affairs and Communities and Local Government)b is responsible for the scheme, but local police and fire authorities administer the schemes.

The Secretary of State for Communities and Local Government (or equivalent for Devolved Administrations)c is responsible for the scheme. Each local authority responsible for a Fund is required to appoint an officer responsible for pension administration. The majority of Funds have a pensions committee overseeing investments.

Source: IPSPC.

a There are equivalent arrangements for schemes run by Devolved Administrations. In Northern Ireland the Department

of Finance and Personnel appoints the Accounting Officers who are responsible to the Northern Ireland Assembly. In

Scotland, Ministers have devolved responsibility for five public service schemes; the day to day running is delegated to an

appropriate Accountable Officer.

b ibid.

c ibid.

124


6.5 There are a great variety of governance arrangements in the public service pension schemes. Some unfunded schemes have governance groups which look at scheme pressures and reform proposals in detail, while others have bodies for negotiating rule changes which do not involve the same level of detailed scrutiny. These groups and bodies include employee representation. Such governance and negotiating groups can only be advisory on matters of policy as the relevant Secretary of State is responsible for policy and for setting scheme rules and is ultimately accountable to Parliament for policy decisions.

6.6 However, only one unfunded scheme, the Principal Civil Service Pension Scheme (PSCPS), has a formal pension board, responsible for managing the scheme in accordance with its governing legislation and rules and for the stewardship of the resources it consumes.

6.7 The Local Government Pension Scheme (LGPS) has a range of local pension committees which consider various matters, including local discretions under national scheme rules and investments, but these do not have the legal status of trusts. At present their fiduciary duty for the Fund monies is to taxpayers rather than to members and other beneficiaries (as it would be in a trust based scheme) and overall responsibility for the scheme lies with the Secretary of State for Communities and Local Government.2 The Policy Review Group for England and Wales run by the Department for Communities and Local Government (DCLG) considers possible changes to rules.

6.8 This position, for both the unfunded schemes and the funded LGPS, contrasts with trust based funded schemes in the private and public sector. These are required by law to have a board of trustees, usually consisting of a fixed number of members (management; nominees of employees and pensioners; and independents). These are required to have knowledge of their scheme (often involving training), meet regularly and oversee pension administration.

6.9 Ultimately the board of trustees is legally responsible for the operation of the scheme, including effective administration of benefit payments and communications with members, the investment of scheme assets, setting appropriate funding principles and plans to recover deficits and ensuring adequate internal controls are in place. The scheme rules are set out in a trust deed, but the powers to amend this deed and the roles of employer and trustees in making amendments vary considerably from scheme to scheme. While there are valid reasons for the difference between the governance models of the public and private sectors, lessons can be learned from the trustee model.

Establishing good governance

6.10 The current arrangements described above mean that there is sometimes no clear separation of duties between those responsible for policy changes, for the governance of the schemes and the delivery of administration. For example, between those charged with

2 Or equivalent Minister in the Devolved Administrations.

125


setting the scheme’s rules, those who oversee the schemes and their financing and those who implement changes in scheme rules and handle other administrative tasks.

6.11 Some of these responsibilities fall to departments running schemes, some to employers and some to contractors. This can lead to a lack of transparency and clarity (for members and the public) as to who is responsible for what, which in turn can confuse lines of responsibility and accountability. To provide greater clarity, avoid conflicts of interest and help increase the focus on efficiency and effectiveness there needs to be clear separation between these roles.

6.12 Evidence presented to the Commission3 (both in the second call for evidence and in stakeholder roundtables) suggested that where there currently are boards, groups or committees, members of public service pension schemes (both funded and unfunded, centrally run and locally administered) are sometimes not formally represented, for example, by nominees specifically elected by the members. In some cases representatives of the workforce covered by the scheme, such as union officials, sit on formal governance or negotiating groups, but schemes vary greatly in this and in the extent to which there is formal member involvement.

6.13 But there are examples of good practice, for example, the majority of local authorities have some form of member representation in their governance arrangements. In November 2008 DCLG issued statutory guidance to local authorities administering pension schemes requiring them to publish governance compliance statements grading themselves against 17 criteria, ranging from the structure of and representation on their pensions committee through to frequency of meetings, voting rights and the training of members. A DCLG survey found that 96 per cent of the 89 local authorities in England and Wales classify themselves as compliant with that guidance. The 2010 annual survey of LGPS funds conducted by the National Association of Pension Funds has found that about 90 per cent of Funds have a LGPS member or a trades union representative on their main pension committee.4

6.14 However, it seems that only a minority of member representatives have full voting rights. UNISON submitted evidence to the Commission that by, 2009, only seven of the 89 England and Wales Fund authorities had allowed voting by scheme members of pension committees. This difference reflects current limitations on Committee members’ roles, for example, where democratically elected councillors are responsible for individual Funds. But there does seem to be some room for improvement in this area; without a standard requirement for member representation members may not feel adequately involved in decisions concerning their pension scheme.

3 See Annex D.

4 NAPF Annual Survey 2010 consisting of 24 Local Government Pension Schemes.

126


Recommendation: Every public service pension scheme (and individual LGPS Fund) should have a properly constituted, trained and competent Pension Board, with member nominees, responsible for meeting good standards of governance, including effective and efficient administration (Recommendation 17a).

6.15 The Commission believes scheme members in all the public services should be able to nominate persons to pension boards and committees along similar lines to the rights of members in the private sector to nominate persons to sit on boards of trustees. Pension boards should therefore include independent professionals and scheme members in similar proportions as apply in the private sector to boards of trustees.5 It is also very important that as well as the ‘lay persons’ there are also independent members, usually professionally trained and with experience of the pensions environment.

6.16 Good board member appointments and behaviours are arguably more important than board structures and much can be learned from good examples, both in the public and private sectors. There will need to be coherent policies on the appointment of members. It will be crucial to the success of pension boards that members have appropriate training and that employers help board members to commit enough time to their duties. Clear guidance will be required for members of pension boards on their role and duties. They would fulfil similar duties to trustees, acting in accordance with scheme rules, impartially and prudently, balancing the interests of scheme beneficiaries and of taxpayers. There will be a need for effective committee structures to facilitate sound decision making and strong oversight of scheme administrators and fund managers.

6.17 It will be important that measures are put in place to ensure clear separation of duties and the avoidance of conflicts of interest by the members of the pension board.

6.18 The Commission recognises that progress has been made in the area of governance structures across the centrally and the locally managed schemes. Two examples of how some improvements have been delivered so far are set out below: the first covers the introduction of a formal pension board to an unfunded scheme, the PCSPS, shown at Box 6.B; the second describes the governance arrangements for a local funded scheme, the London Pension Fund Authority and is shown at Box 6.C.

5 The Pensions Act 2004 stipulates that all trustee administered pension schemes should have a minimum of one-third of

trustees nominated by members of the scheme.

127


Box 6.B: Principal Civil Service Pension Scheme (PCSPS) Governance

The PCSPS covers over 200 employers and has 1.5 million members. Following a review of PCSPS administration, the Cabinet Office has transformed administration arrangements and governance, adopting best practice in the private sector, where possible.

A twelve member Scheme Management Board was set up in April 2010 chaired at Permanent Secretary level with four members’ representatives, four representatives from scheme employers, one representative of central government (HM Treasury) and two (non-voting) non-executive members who chair the Board’s Risk and Operations Committees. All Board members were selected on the basis of having relevant experience in areas such as finance and human resources.

The Board’s terms of reference specify its role as:

• managing the Scheme in accordance with the relevant legislation and rules;

• developing and managing a risk management framework and internal controls system; and

• oversight of the Scheme administrator.

As Scheme Manager (with responsibilities akin to a trustee board for a private sector scheme) the Board reports to the Head of the Civil Service, who is the Accounting Officer and is in turn responsible to the Secretary of State and ultimately to Parliament. The Board oversees improved service for members while at the same time providing reassurance to employers, and ultimately the taxpayers who fund those employers, that the Scheme’s administration gives value for money.

The Board is supported by an executive that manages relationships with the Scheme administrators and the employers covered by the Scheme. The executive has participation agreements in place with each employer, which set out their responsibility to provide accurate data and the right financial contributions to the Scheme.

128


Box 6.C: London Pension Fund Authority (LPFA) Pension Board

The LPFA is a statutory body, with a local authority’s powers in relation to the LGPS, although unlike almost all other LGPS Funds it has been set up independent of direct local political control. It has over 200 employing authorities and 73,000 members. The LPFA has a constitutional document setting out a formal governance structure of its pension board and committees.

The LPFA Pension Board Chairman and Vice Chairman are appointed by the Mayor of London and are independent of the employers covered by the fund. The Mayor appoints a further nine board members following public advertisement, five of which are subject to consultation with representatives of London local government. The aim is to have a mix of board members with a variety of skills including investment management, business, pensions finance, local government finance, general management and corporate social responsibility.

The Board regularly meets and reviews its strategic objectives for the medium term, including the investment strategy. The Board delegates business to its committees which review investments, performance, administration and auditing. In addition, a remuneration committee and an urgency committee meet on an ad hoc basis as required. For example, the Board sets the overall investment policy and strategy of the Fund and the investment committee is responsible for implementation, including the appointment of Fund managers and monitoring the performance of the Fund and of investment managers against targets.

Regular newsletters are sent to all members and all employers. Annual forums are held for Fund members and employers. A Fund member panel has been established and is invited to attend each board meeting. A formal panel meeting is held after the Board with board members and senior officers in attendance. There is also an employer panel with similar arrangements.

6.19 The introduction of new governance arrangements would not diminish the role of the responsible government minister who will remain legally responsible and in overall control of each scheme, or that of the formally appointed accounting officers for schemes, nor, for LGPS Funds, would it replace the statutory administering authority6 responsible for implementing ministers’ policies and regulations. This reflects the point that these pension promises are backed by the State, and therefore taxpayers, and that the legislative and executive arms of government should retain overall responsibility for public finance and expenditure.

6.20 A diagram showing the responsibilities that the public service pension boards would be expected to perform is shown at Box 6.D.

6 The Fund is usually controlled by locally elected councillors.

129


Box 6.D: Proposed governance arrangements for scheme administration

Source: IPSPC.

Pensions Board

• enforcing the rules

• oversight of scheme administration

• appointment of scheme actuary and auditor

• risk management and internal controls

• oversight of financial management

• investement management (LGPS only)

• oversight of appeals

Sch

eme

rule

s &

leg

isla

tio

n

Ind

epen

den

t o

vers

igh

t

Investment advice

(LGPS only)

Scheme Administrator

maintenance of records, collection of contributions, calculation of benefits,

communications with employers and members, publication of data, financial management,

processing appeals, ill health benefits

Employers

Data & contributions

Guidance on roles

Del

iver

y o

f b

enefi

ts &

co

mm

un

icat

ion

s

Scheme members

External advisors

(auditors, scheme

actuaries)

130


6.21 The Pension Board would focus on the implications of administering pension scheme rules. However, even if all schemes have a pension board in future, there will still be a need for separate pension policy groups to consider at national level major changes to scheme rules and the value of the pension scheme to the membership. As noted in paragraph 6.5 above, many schemes have such groups or bodies.7 These tend to be established as part of the consultation and negotiation machinery for handling pensions as an element of the remuneration package and to have member and employer representation as appropriate.8 The Government should introduce such groups for the schemes that do not have them. All policy review groups should ensure that information about key proposals for change and related costs is publicly available.

Recommendation: There should also be a pension policy group for each scheme at national level, for considering major changes to scheme rules (Recommendation 17b).

Locally administered schemes

6.22 The Commission does not propose rationalising governance and administration of local schemes through moving to wholly national arrangements, instead advocating greater co-ordination and collaboration consistent with retaining local identity and accountability. There would be some efficiency to be gained from putting the locally run schemes (the LGPS, police and firefighters schemes) fully on a national basis, so that they were run in the way that the scheme for locally-employed teachers in England and Wales is run. However, the Commission recognises that such a potential change raises issues to do with local accountability and the role of democratically elected local government representatives, which goes beyond consideration of good governance for pension schemes. Equivalent considerations apply to the separate schemes run by Devolved Administrations, such as the schemes for NHS staff in Scotland and in Northern Ireland.

6.23 Neither does the Commission advocate a uniform governance model in terms of the way responsibilities are allocated within the updated governance structures. The Commission has been given examples of best practice in pension scheme governance and evidence of how different schemes have established their governance structures and it is clear that a variety of different models operate across the public service schemes to good effect.

6.24 However, where schemes with nationally determined rules are administered locally, as is the case for the LGPS, police and firefighters schemes, there is a case for supplementing local pension boards with a national pension board for each scheme separate from the individual local authorities and employers. The locally administered schemes are a hybrid of

7 There are, for example, such groups for most of the very big schemes: NHS and Civil Service Pension Scheme

Governance Groups; a Teachers Pensions Committee; the Police Negotiating Body; the Fire Pensions Committee; and the

LGPS Policy Review Group.

8 In some cases these groups operate separately from the remuneration negotiations between employer representatives

and members.

131


nationally set statutory scheme rules, subject to a minor amount of local discretion and local management.

6.25 It would be for the Government to decide how they would be set up, for example, whether this would be under the Department responsible for the scheme’s rules or another body. These national pensions boards might build on existing arrangements such as operated by the Local Government Association and the Local Government Employers organisation for the LGPS in England and Wales.

6.26 The role of national pensions boards in respect of the locally administered schemes would be facilitative. While not taking on a regulatory or oversight role national pensions boards might help analyse and recommend on comparative performance of governance and administration. They might point to the scope for efficiencies in those areas that raise standards across the piece. In the case of the LGPS, their national pensions board could also look at overall funding and the comparative performance of individual Funds, fostering links between different individual LGPS Funds and perhaps making recommendations on opportunities to capture efficiencies and to improve investment performance. The boards could include nominees of individual scheme members and employers as well as those responsible for national scheme rules and local financing.

Transparency

Communication with scheme members

6.27 Not all schemes communicate with active members on a regular basis. Currently defined contribution schemes are required to provide members with an annual benefit statement which shows a statutory money purchase illustration, based on a number of assumptions, projecting a possible income in retirement. However, almost all public service pension schemes are defined benefit and defined benefit schemes are formally required to provide a statement only if the member requests one.9

6.28 Some public service schemes do provide annual statements that may show information such as a projected annual pension and lump sum if the member retires10 at their Normal Pension Age (NPA), or had died or retired on ill health grounds, or otherwise took a pension around the date the statement was produced. But some schemes do not provide regular statements to members.

6.29 The Commission considers that regular updates to scheme members regarding pension benefits earned to date and forward projections of pensions are crucial in improving general pension education and in promoting a sense of personal ownership of pension benefits.

9 The Personal Pension Schemes (Disclosure of Information) Regulations 1987 and the Occupational Pension Schemes

(Disclosure of information) Regulations 1996.

10 This may be based on same level of pensionable pay that would be used to calculate an immediate pension award but

assuming that level continues in future.

132


Statements could also usefully include member and employer contribution rates. The information in statements should be clear and as easy to understand as possible.

Recommendation: All public service pension schemes should issue regular benefit statements to active scheme members, at least annually and without being requested and promote the use of information technology (IT) for providing information to members and employers (Recommendation 18).

6.30 Some public service pension schemes provide pension calculators which allow scheme members to access information via web-sites. Such calculators can be used to provide estimates of their pensions at specific career or age points, which can, for example, be useful when someone is considering retiring early or working beyond their normal pension age.

6.31 The Commission welcomes the increasing use of technology, such as web-sites, the provision of online pension calculators and the provision of annual statements electronically to members if they wish. It should become standard to utilise a variety of channels to communicate pension scheme information to members and employers, in addition to paper-based methods. Government should promote the use of IT to assist members in understanding their pension entitlements and retirement options.

6.32 Those running individual LGPS Funds should also provide details of Fund investments and performance to scheme members.

Publication of scheme data

6.33 Scheme members, the public, Parliament11 and commentators should be able to access scheme data easily to enable them to determine the performance, viability and key facts associated with the different schemes. However, the Commission has concluded that at present the availability of such data is at best patchy: some key data is not available, at least not publicly.12 This needs to be improved.

6.34 Currently the various schemes, local administrators and others such as HM Treasury, the Office for National Statistics (ONS) and the Office for Budget Responsibility (OBR), publish different information in reports, accounts, valuations and statistical compilations. Departments responsible for individual schemes and also individual local authorities do publish reports containing some information on public web-sites. However, there is no central, publicly available, depository of information to enable comparisons between schemes or individual administrators, e.g. between administration costs, membership profiles and (in respect of the LGPS) return on investments.

11 Including Parliaments and Assemblies for the Devolved Administrations.


133


6.35 The Government does not at present publish a single set of figures that covers the total fiscal impact of public service pension schemes, but instead publishes partial figures on differing bases in a variety of places. Equally government does not identify separately the total amount of public spending on the LGPS, police or firefighters scheme at national level, as, for example, it does with the NHS. This lack of transparency prevents comparisons and hinders adequate analysis, which adversely affects the quality of the public debate about the future of public service pensions.

6.36 There should be ready access to information which demonstrates that a scheme is being well-managed in accordance with the relevant rules and regulations, that it is fully compliant with all codes of practice and with other requirements. Full valuation reports as well as annual accounts should be published. For the police, firefighters and the LGPS, these should cover the position nationally as well as locally and enable comparisons across different authorities, for example, between particular aspects of the individual LGPS Funds. For funded and unfunded schemes they should also separate government employers and admitted bodies and show both the potential membership and actual membership.

6.37 In the case of the LGPS, this data also needs to include consolidations across all LGPS Funds that enable investment performance to be compared over time. Where schemes or LGPS Funds necessarily do things differently, either in comparison to each other or to private schemes, then this should be explained accordingly.

6.38 Similarly detailed membership data (consistent with data protection requirements) might be released regularly and related to the value of prospective and actual pension benefits payable.

6.39 Without such overall data on the schemes it is not possible to see the extent of challenges and the potential costs of the benefits accruing.

Recommendation: All public service pension schemes should regularly publish data which, as far as possible, is produced to common standards and methodologies and is then collated centrally. This information should be of a quality that allows simple comparisons to be made across Government, between schemes and between individual LGPS Funds (Recommendation 6).

Monitoring of ill-health retirement

6.40 The Commission has seen evidence13 which illustrates that changes in the monitoring and design of ill-health pension provisions and measures to avoid and remedy the causes of ill-health retirement have led to major reductions in ill-health retirement levels across schemes. Examples include: local authority-maintained schools in England and Wales, where levels fell from 18 per cent of all retirements in 1997-98 to 3 per cent in 2008-09; and the

13 Information provided by schemes.

134


NHS, where ill-health retirements in England and Wales fell from around 23 per cent of all retirements in the mid to late 1990s to around 8 per cent in 2009-10.

6.41 In the case of police and firefighters, in the late 1990s ill-health retirements represented about half of all annual police retirements (13 per 1,000 active members) and about two-thirds of all annual firefighter retirements (19 per 1,000), whereas the present figures for both are between 2 and 3 per 1,000 actives.14 This is a welcome development. However, the Commission noted that data, while available, is fragmented and not always comparable.

6.42 As part of improved transparency of data, the Commission suggests that ill-health retirement figures in all public service pension schemes should be collected and published on a regular and common basis. This would make it easier for government to systematically compare ill-health retirement rates across the major schemes.

External scrutiny

6.43 The Commission believes that improved governance, and transparency and accessibility of key data, would be assisted by greater external and independent scrutiny of public service pension schemes.

6.44 Some form of external scrutiny of public service pension schemes would help build trust and confidence of both the members and taxpayers in the running of public sector schemes by encouraging schemes to follow common and appropriate standards of governance and administration, publish appropriate data and provide an overview of their performance.

6.45 Currently the departments which sponsor particular schemes, such as DCLG for the local government and firefighters schemes, Education for the teachers, the Home Office for police and Health for the NHS, together with HM Treasury as the lead department for public service pensions policy, combine to oversee the performance of public service pension schemes. The argument has been put to the Commission that the statutory backing for the schemes effectively guarantees that scheme members’ benefits will be paid and therefore external scrutiny of public service schemes does not need to be as extensive as The Pensions Regulator’s functions in respect of private sector schemes. The Commission agrees with that.

6.46 However, even with this government guarantee to meet the pension promise, there is a case for scrutiny of public service pension schemes that is independent of stakeholders with a direct interest, such as employers, local councillors and ministers. This scrutiny should cover issues such as the service provided to scheme members and the delivery of value for money for taxpayers. While some local authorities consider the DCLG as their pension regulator, this does not seem appropriate given that best practice advocates the separation of powers. Audit Scotland has stated that they would not regard the Scottish Public Pensions Agency

14 Information provided by schemes.

135


(the DCLG’s equivalent in Scotland) as regulators as they have no oversight of pension administration or the management of pension funds in the administering authorities.15

Recommendation: Governance and the availability and transparency of information would be improved by government establishing a framework that ensures independent oversight of the governance, administration and data transparency of public service pension schemes. Government should consider which body or bodies, including for example, The Pensions Regulator, is most suitable to perform this role (Recommendation 19).

6.47 This framework should include the requirement for all schemes to meet minimum standards of governance and administration. This independent external oversight would not replace existing internal and external audit of schemes, including by bodies such as the National Audit Office and Audit Scotland.

Long term sustainability

6.48 When it comes to assessing the viability of continuing to provide the public service pension schemes in future and of financing the cost of past pensions promises, there is a need for fiscal policy to take account of the sustainability of such commitments over the long term, looking at the schemes’ long term impact on public finances in the context of other pressures on public spending.

6.49 Chapter 4 discusses the concept of an overall cost control mechanism, and recommends that a cost ceiling be implemented based on measuring cost as a percentage of pensionable pay. This cost control is based on the value of benefits accruing within a particular period. In addition to monitoring and controlling costs on this basis, it will also be important to monitor and make transparent the commitments that have been made and the likely implications of these in the future.

6.50 There has, in the last few years, been a focus on net cash movements for the unfunded schemes which looks at how much the Exchequer has had to pay centrally to cover pension expenditure after allowing for employer and employee pension contributions. However, as the interim report noted16, this is an inherently volatile measure, reflecting changes in the balance between active members who contribute to schemes and pensioner members who draw benefits. Looking at net cash movements therefore obscures the point that most of the contributions come from public monies and the obligations have to be met irrespective of whether or not contributions are levied by schemes. As discussed in Chapter 5, it is reasonable for government to continue to structure public service pension provision on a mainly unfunded basis, but this needs to be done in a fully transparent manner and, as noted in paragraph 6.36 above, some improvements could be made.

15 Audit Scotland (2011) The cost of public sector pensions in Scotland.


136


6.51 For the funded LGPS, it is arguably easier to assess whether the pension benefits promised are sustainable by considering whether the liabilities built up to date are being fully covered by the assets held by the Funds and any deficit recovery plans approved by the Fund. Relevant information on assets, liabilities and deficits for individual LGPS Funds is generated by triennial actuarial funding valuations. However, this information is not available at an aggregate national level.17 Although the overall position of the LGPS is due to feature in Whole of Government Accounts, it is the funding valuations, not accounting data, which affect LGPS contribution rates and council tax. This means that, at present there is no independent and publicly available assessment of the likelihood of the LGPS in England and Wales eliminating its overall deficit over the long term and the consequences of not doing so. That overall deficit is in effect an unfunded liability.

6.52 The Audit Commission made an initial examination of the sustainability of the LGPS in its July 2010 information paper, ‘Local government pensions in England’18 and took the view that the current approach, with unfunded liabilities being deferred into the future, could not continue indefinitely. Following the planned abolition of the Audit Commission, it will be necessary for another body to follow up this conclusion.

6.53 Two measures that could be used to examine sustainability in the public service pension schemes are: a measure based on an estimate of total liabilities; or one based on a projection of actual gross cash payments to pensioners. The latter option would need to consider how, in funded schemes, payments are made out of the individual Funds and so the link to Government expenditure is indirect. A liabilities measure represents the value in today’s money of all expected future pension payments from benefits that have been built up to date. Such measures already exist for schemes and will feed in to Whole of Government Accounts.

6.54 A further version of liabilities for the unfunded schemes is in the process of being worked up by the ONS, as part of the updating of the European System of Accounts (ESA) within the worldwide System of National Accounts that is due to come into effect from 2014.19 The UK Government has also been conducting a consultation on the discount rate they use for funding valuations, as recommended by the Commission in its interim report.20 For a liabilities-based measure of long term sustainability, the OBR might consider producing an analysis that includes the discount rate determined following the current UK Government consultation as well as other potential rates, including the rate eventually adopted for the new ESA. However, the interim report21 concluded that a liabilities measure is of limited use when assessing sustainability.

17 The individual valuation reports are sent to the responsible Minister, so his advisers can check they meet the

requirements set out in scheme regulations.

18 www.audit-commission.gov.uk/localgov/nationalstudies/localgovpensions/Pages/Default_copy.aspx.

19 There has already been a decision in principle by ESA members to use a long-term fixed discount rate to value the

pension liabilities, rather than using the annually revised and hence volatile discount rate employed in accounting for

such pensions, although there has been no decision as yet about the final discount rate to be used. This approach on the

discount rate is similar to that used now for funding valuations of the unfunded public service schemes.

20 IPSPC (2010) Interim report, paragraphs 4.64-4.66.


137


6.55 Alternatively, the OBR could seek to build on the long term projections of future cash payments to pensioners in the unfunded schemes, which were published from 2004 to 2009. These plot projected cash payments to pensioners as a percentage of projected nominal Gross Domestic Product (GDP), using GDP as a proxy for the size of the tax base available to finance those payments. In its interim report the Commission argued that looking at projected public service pension benefit payments of the unfunded schemes as a percentage of GDP is an effective measure of the future cost of those schemes.22

6.56 As the LGPS is a funded scheme, where benefits are paid from the individual Funds, equivalent projections of future payments to pensioners would not represent the impact on public finances. Alternative measures would be needed. These might include projected long-term levels of employer pension contributions, made by the public sector employers within the LGPS.23

6.57 Projected cash flows, such as described above, can, for example, capture the potential impact of changes in the size and composition of the future workforces covered by schemes and can show the direct and indirect fiscal impact. If a liabilities measure were preferred, or used to supplement un-discounted figures, the Commission would advise against using an annually variable or otherwise volatile discount rate rather than a long term fixed rate. However, it would be for the OBR to determine how sustainability should be assessed and it need not confine itself to one approach.

6.58 The OBR needs to be empowered to request from appropriate sources whatever data it considers necessary to discharge this responsibility.

Recommendation: When assessing the long term sustainability of the public finances, the Office for Budget Responsibility should provide a regular published analysis of the long term fiscal impact of the main public service pension schemes (including the funded LGPS) (Recommendation 20).

Investment of local government pension monies

6.59 Managing investment funds is an additional aspect of pension scheme governance for the funded public service pension schemes. As the Audit Commission has noted, maintaining the financial health of each individual LGPS Fund is a complex issue that involves having to balance competing requirements that change over time against the background of statutory backing for the LGPS. The admission of thousands of non-local authority private sector employers to the scheme adds a further complication as their solvency is not backed by taxpayers.


23 In projecting employer pension contributions it would be necessary to allow for factors such as the impact of LGPS

Funds’ average investment returns and levels of employee contributions.

138


6.60 At present and in line with DCLG guidance, each individual LGPS Fund follows a Fund-specific strategy which sets out how its pension liabilities are best met, in terms of investments and reasonable risk parameters. This is balanced against other aims, such as seeking to keep employer contribution levels stable and affordable and taking account of the impact on local council tax payers.

6.61 However, the LGPS Funds are given far more scope to exercise flexibility over their approach to funding than trust-based schemes, where the regulatory regime accepts that funding should be scheme-specific but triggers firm criteria for funding adequacy and ensures that schemes take appropriate action. The LGPS approach helps in balancing conflicting aims. It reflects the Funds’ ability to accept higher levels of investment risk and their capacity to recover deficits over decades. These in turn arise as a result of the statutory basis of the LGPS, the constitutional permanence of local government as an employer and a Fund’s current positive cash flow position.

6.62 In seeking to make best use of the scope conferred by their status, the individual LGPS Funds are in many respects models of good practice in their governance of funding and investment. In general, they publish openly and follow transparently a clearly stated set of relevant policies, covering key areas, such as: setting appropriate investment objectives, identifying and controlling risks and communicating with stakeholders. They typically review all aspects of these policies and of performance in line with them, on a regular basis, taking action in the light of review findings to try to ensure that the standard of the governance arrangements is maintained and where possible improved.

6.63 However, Funds could in theory gradually build up ever larger deficits, in effect accepting higher long-term costs in order to secure short term affordability for employers and employees. The Audit Commission cited evidence to the Commission that this has been happening.24 The Audit Commission also indicated in the same report how investment strategies are vulnerable to a significant reduction in the proportion of active members, which could result in a negative cash flow.25 That would mean that pensions could not be paid without cashing in investments, which could reduce investment in long-term growth assets.26

6.64 Neither of these important points, which clearly relate to the long term sustainability of LGPS funding, is readily apparent from the large volume of information about funding strategies and investment performance currently published by and about LGPS Funds. This suggests that the management of funding in the LGPS is not in some critical respects sufficiently transparent.

24 Audit Commission (2010), Local government pensions in England’ paragraph 43, www.audit-commission.gov.uk/

localgov/nationalstudies/localgovpensions/Pages/Default_copy.aspx.

25 As explained in the interim report paragraph 4.79.

26 See paragraph 31 and footnote in ‘Local Government Pensions in England’. This negative cashflow would arise in any

case in due course as the LGPS become a more mature scheme, with an increasing proportion of the liabilities relating to

pensioners rather than active members.

139


6.65 There is a strong case for making sure that ministers, Parliament, commentators and taxpayers can understand the trends regarding the funding of liabilities, for example, whether or not there is an established or developing trend to defer unfunded liabilities into the future. This is a good indicator of whether the amount being invested, after taking into account the unique characteristics of LGPS Funds, is sufficient to meet liabilities in the long term. This should supplement, not replace, the existing legal responsibilities of individual Funds, including for publishing information locally about the Fund.

6.66 Similarly, it is important to make clear how much is being added to long term costs by decisions to recover deficits over a period longer than a standard benchmark term. That would be a measure of how far future generations are facing greater burdens because of decisions to trade off the short term against the long term. Such information cannot easily be obtained at present for individual Funds or for the LGPS collectively. There is also a gap in the availability of regular information about the comparative investment performance of different Funds, as discussed in the section in this chapter on publication of scheme data.

6.67 The Audit Commission and others have also argued in their evidence to the Commission that significant improvements in investment performance and risk control could be generated in a number of cases, in particular across London, by moving to fewer, larger LGPS Funds. However, the way that investment arrangements have been linked to each individual Fund’s governance arrangements seems to have been an obstacle. There is, for example, a risk that potential investment gains are seen as outweighed by loss of influence over employer contribution costs. An individual LGPS Fund might not pursue rationalisation through pooling investment assets with other Funds if it then expects to have to alter assumptions about investment growth or to change deficit recovery periods in ways that put upward pressure on contribution costs. It would be desirable for LGPS Funds to have incentives to obtain performance improvements, including merging the investment of assets or even the underlying Funds where appropriate.

6.68 The governance of LGPS funding and investment at present seems to be too narrowly focused on the individual Fund rather than the overall sustainability of the LGPS. Therefore, as recommended earlier in this chapter, comprehensive data covering all LGPS Funds should be published. In addition, the OBR should include the LGPS in looking at the long-term sustainability of public service pensions.

Recommendation: Centrally collated comprehensive data, covering all LGPS Funds, should be published including Fund comparisons, which, for example, clarify and compare key assumptions about investment growth and differences in deficit recovery plans (Recommendation 21).

140


Improvements to administration and rationalisation of costs6.69 Good administration is essential to the delivery of accurate and timely pension payments. However, good pension scheme administration goes beyond this, including accurate record keeping and the provision of appropriate and timely disclosure to all scheme members, as well as to the individual employers in multi-employer schemes. This may be in the form of guidance manuals, individual benefits statements, the provision of online calculators, or the operation of a call-centre to provide members with one-to-one advice and query resolution.

6.70 Expenditure on the administration of pensions is a significant cost to schemes: as an example, in 2009-10 the LGPS in England paid £115 million in administration costs (not including investment management).27

6.71 A great deal of evidence was received by the Commission regarding the costs of the administration of pensions and overall the majority of the larger public service pension schemes compare very favourably with the average private sector cost (among the largest schemes) of £41to 47 per member.28 By way of comparison, the average LGPS Fund administrative costs were in the order of £2429 to £28.30 For the larger unfunded public service pension schemes, annual administrative costs per scheme member (active, deferred and pensioner) range from £631 for the Teachers Pension Scheme (TPS), through £1632 for the NHS Pension Scheme (NHSPS), to £2433 for the PCSPS.

6.72 It is important to avoid drawing superficial conclusions about differences between schemes’ administration costs as different schemes have different characteristics and offer different levels of service to their members, including for example in the provision of benefit statements. Average costs per member will also vary significantly depending upon whether deferred members are included in the calculations; particularly if little or no cost is expended on this, often significant, percentage of the scheme membership. Also important are membership characteristics such as staff turnover, ratios of deferred members to actives and pensioners and members’ IT literacy and access; and, of course, the overall size of the schemes and the numbers handled by individual scheme administrators within each scheme, which vary considerably.

27 Local Government Pension Scheme Funds England 2009-10, Statistical Release, 13 October 2010, Communities and

Local Government, www.communities.gov.uk/documents/statistics/pdf/1738358.pdf.

28 Capita Hartshead Annual Pension Administration Survey 2010; the average administration charges were £47 per

member with in-house administration and £41 with outsourced administration.

29 The National Association of Pension Fund 2009 Annual Survey.

30 Chartered Institute of Public Finance & Accountancy Benchmarking Club, which covers 68 LGPS local authorities across

the UK.

31 Information provided by the Scheme to the Commission.

32 ibid.

33 ibid.

http://www.communities.gov.uk/documents/statistics/pdf/1738358.pdf

141


6.73 Evidence submitted to the Commission did not by itself demonstrate either good or bad public service pension scheme administration, but it did highlight the absence of a clear definition of what good standards of administration and governance might be.

Recommendation: Government should set what good standards of administration should consist of in the public service pension schemes based on independent expert advice. The Pensions Regulator might have a role, building on its objective to promote good administration. (Recommendation 22a).

6.74 Many schemes already submit themselves to benchmarking assessments, but participation is inconsistent and voluntary. Schemes should be required to participate so that value for money comparisons can be made, examples of best practice can be identified and changes can then implemented.

Recommendation: A benchmarking exercise should be conducted across all the schemes to help raise standards where appropriate (Recommendation 22b).

6.75 The Commission has received suggestions and evidence from a number of commentators that public service pension schemes offer scope for streamlining and combining of their administration functions.

6.76 Looking first at the administration of locally administered schemes, the obvious question is whether it is efficient and desirable to have over 50 police authorities and perhaps over 50 fire authorities and 101 different local bodies administering the LGPS (89 in England and Wales, 11 in Scotland and one in Northern Ireland) when they are schemes that essentially have nationally set rules with only minor discretionary local variations. There seems an opportunity for these schemes to combine their operations to reduce costs and potentially exploit economies of scale.

6.77 There is clear evidence that the administration of pension schemes can benefit from economies of scale, particularly where existing schemes are below 100,000 members. Chart 6.A demonstrates how economies of scale can dramatically reduce costs of administration. The data is based on 90 defined benefit schemes with a total of 37 million members within the UK, Europe, USA, Canada and Australia collected during 2009 and 2010. The largest had 4 million members and the smallest 10,000. The UK part of the sample included 15 schemes, both public and private sector, covering 6 million members.

142


Chart 6.A: Economies of scale in pensions administration

£0

£10

£20

£30

£40

£50

£60

0 100,000 200,000 300,000 400,000 500,000 600,000 700,000 800,000 900,000 1,000,000

Total Membership (active, deferred and pensioners)

Adm

inis

trat

ion

Cos

t Per

Mem

ber

Per

Ann

um –

Log

arit

hmic

Sca

le

Source: CEM Benchmarking Inc., 2011.

6.78 The Commission welcomes the fact that a number of local authorities, particularly in the LGPS, have already begun to explore opportunities to share administrative services and contracts (the Commission received evidence about existing or potential projects in Scotland, Wales, the South-West of England and London). The costs and risks for local authorities from such ventures will need to be assessed carefully, but at this early stage it would seem reasonable to assume that there will be financial savings.

Recommendation: Central and local government should closely monitor the benefits associated with the current co-operative projects within the LGPS with a view to encouraging the extension of this approach, if appropriate, across all local authorities (Recommendation 23a).

6.79 Opportunities to share services and contracts between the unfunded centrally administered public service pension schemes appear less obvious.

6.80 Different unfunded schemes have a variety of scheme features, for example, normal pension ages, contribution rates (both employer and employee), accrual rates and benefit packages. This makes it more difficult to share administration. However, this Commission’s proposals will reduce the differences between scheme designs.

143


6.81 Also there has already been some movement to streamline administrative arrangements for some of the major schemes that deal with multiple employers. For example the introduction of the My Civil Service Pension programme34 in the civil service has accelerated the process of reducing the number of civil service pension scheme service centres from ten to six. In addition, much of the cost of payroll and benefits administration is already outsourced. For example, the Ministry of Defence estimates that outsourcing represents some 80 per cent of the costs of administering its pensions.

6.82 Nevertheless, there may be opportunities for greater use of shared services. But any potential savings and efficiencies from sharing by schemes must be weighed against the risks associated with integration issues, one-off costs and operational challenges given the difference in rules and current administration procedures between these schemes.

Recommendation: Government should examine closely the potential for the unfunded public service pension schemes to realise greater efficiencies in the administration of pensions by sharing contracts and combining support services, including considering outsourcing (Recommendation 23b).

6.83 Other aspects of public service pension administration also have to be considered. There are currently a multitude of public service payroll and pension administrative systems in operation, using a variety of different IT platforms, software and data formats. There is a great variety of scheme rules: most schemes need to operate systems for members being pensioned under current rules and for members still being pensioned under old rules. At best having a multitude of arrangements can impede efficient administration and at worst potentially cause errors and inconsistencies and make it difficult to implement reform.

6.84 The Commission has also heard evidence of monthly payroll details necessary for pension calculations, such as contribution amounts, being regularly transferred late to outsourced contractors, with resultant delays in payments. The public service pension schemes seem to have lagged behind the private sector with regards to investment in payroll and pension IT systems. Scheme administrators have emphasised to the Commission that there is considerable pressure on their systems because of the various changes they are already managing. This situation will need to be taken into account when considering implementation of the reforms.

The case for a new legal framework

6.85 The existing public service pension schemes are established under numerous, separate, sources of legislation, and in the case of the Armed Forces Pension Scheme 1975 under Royal Prerogative. There is no single legislative framework to cover issues such as:

34 A programme to improve the governance and administration of the Civil Service pension schemes.

144


• the basis on which the schemes can be amended;

• the procedures to be followed when implementing any such amendment, including relevant bodies to be consulted and the nature of such consultation in respect of a proposed change;

• what constitutes an accrued right and the extent to which such rights are protected;

• the structure of governance within the schemes; and

• defining where responsibilities in respect of the schemes rest.

6.86 The current position means that some schemes may be amended by consultation with appropriate associations and representative bodies,35 whilst other schemes contain specific consent requirements. Also, some schemes can be amended by secondary legislation, whereas others, in particular judicial pensions, require primary legislation for all but very minor amendments. This clearly impacts on the flexibility of the schemes where procedural changes are necessary, making them compatible with provisions for occupational pensions in general under UK and EU law, and can be seen to create a degree of unfairness across the public service schemes.

6.87 Although the legislation which currently establishes the existing public service pension schemes (such as the Superannuation Act 1972) typically provides a general power to implement pension benefits, such power is conferred on many people. For example, the power to implement and administer the PCSPS is with the Minister for the Civil Service, whereas the equivalent power in respect of the LGPS, TPS and NHSPS rests with the relevant Secretary of State (although any change to the TPS and NHSPS regulations is ultimately subject to the consent of HM Treasury).

6.88 The current primary legislation, other than for judicial pensions, does not provide any detail on issues such as the nature of the scheme and benefits to be provided, the manner in which the scheme should be governed or the division of responsibilities among those responsible for implementing and managing the schemes. To date, the general enabling powers contained in legislation such as the Superannuation Act 1972 have been relied upon to introduce all provisions necessary for the administration and management of the schemes.36 However, in some schemes there are currently no such express provisions under which regulations could determine administration and management obligations and responsibilities. Overall this leads to a lack of transparency in terms of who is responsible for the schemes and how decisions affecting those schemes can and should be taken.

35 For example, the LGPS can, at present, be amended for existing members in respect of their future service provided the

Secretary of State has consulted with “such associations of local authorities as appear to him to be concerned, any local

authority with whom consultation appears desirable and such representatives of other persons likely to be affected as

appear to him as appropriate.” (Section 7(5) of the Superannuation Act 1972).

36 For example, the Local Government Pension Scheme (Administration) Regulations 2008.

145


The benefits of change

6.89 There appears to be a strong case for introducing new, overarching, primary legislation to establish the new public service pension scheme framework. This would provide greater transparency, simplicity and certainty over the reforms and ensure they satisfy common basic principles.

6.90 New primary legislation would need to cover key areas, such as:

• who has ultimate responsibility for the new schemes;

• the regulatory framework;

• the governance structure;

• provisions dealing with scheme amendments, including consultation obligations;

• the extent to which member benefits are protected;

• member involvement and engagement, including communication requirements; and

• transitional provisions in respect of existing schemes.

6.91 In particular, new primary legislation could be structured to:

• provide for a standard power of amendment to be used across all schemes. Such power of amendment could determine the nature and extent of consultation as well as any formal consent needed for future changes. This might address the overall role of HM Treasury in respect of controlling schemes and public service pensions policy and establish whether Treasury consent would be required for future changes in benefit design as well as for key assumptions for valuing benefits in order to help maintain the common UK public service pensions framework. Providing such decision-making or ultimate consent, power to HM Treasury would certainly be one option for consideration given the need to ensure cost control for the taxpayer. It is notable in this respect that amendments to regulations for schemes such as the NHS and the TPS are already, ultimately, subject to the formal consent of HM Treasury;

• confirm the level of protection afforded to members’ accrued rights and define what constitutes an accrued right. In particular, such a definition could clarify the extent to which any protection is afforded beyond the benefits which a member would receive as an ‘early leaver’;

• establish the core benefit design features of the new schemes and any scope for variation at scheme level and the procedures for implementing such variations;

146


• provide the basis on which the cost ceiling for the new arrangements is determined, reviewed and modified. This would include determining who or which body would have responsibility for this and what factors would be taken into account in the timing and extent of any review;

• create a new body, or appoint an existing body, to have overall responsibility for oversight of public service pension schemes and the quality and transparency of information about them. Legislation could establish the statutory duties of such a body, its powers and relationship with the schemes and the employers participating in them;

• establish the framework for a formal governance structure for the new public service pension arrangements; and

• introduce transitional provisions which would apply in respect of the benefits provided by the current schemes.

6.92 The protections might also cover the extent to which there might be limitations on adjustments to existing judicial pensions to meet international conventions for protecting judicial remuneration, while also having regard to factors such as increases in the value of pensions from increasing longevity.

6.93 The new primary legislation would need to be supplemented by regulations containing the detail on matters such as benefit structure, regulatory framework, reporting requirements, disclosure obligations and member participation on appropriate committees.

6.94 Consideration would have to be given to structuring any such legislation in a way which covers those numerous public service pension schemes which have been established by or for the staff and members of non-departmental public bodies (NDPB), often under general statutory powers used to establish those bodies. It will therefore be necessary to clarify which NDPB schemes are intended to be covered by the new legislation, whilst recognising that there may be some schemes (such as those that were established under a separate trust) that cannot be amended directly by legislation.

6.95 The introduction of any overarching legal framework for UK public service pensions would also have to recognise the issue of devolution. Although pension policy is not, currently, a devolved function, it would nevertheless be necessary to consider the role of devolved administrations within the national pensions policy framework discussed in Chapter 5 and how they would implement pension provisions within that overarching framework.

Recommendation: The Government should introduce primary legislation to adopt a new common UK legal framework for public service schemes (Recommendation 24).

147


7 Delivering the change

Box 7.A: Summary

• Implementation is key: the first stage of reform will involve the more detailed development of proposals for applying the common design principles, which will need a structured approach with central co-ordination, for example, to set cost ceilings and timetables for consultation and overall implementation.

• However, consultation on details should be undertaken inclusively, scheme by scheme. This will give employees and their representatives the opportunity to participate fully in the detailed design of the schemes, with the statutory consultation only being carried out once this is complete. Each statutory consultation will need a full equality impact assessment to allow for deeper consideration of the issues for various groups.

• This approach will also allow a longer timeframe for implementation, where this might be needed for specific schemes because of the scale of the change or for legal reasons, for example, for schemes such as the armed forces, police and judges.

• There will also need to be early upfront communication with members, to encourage participation in the consultation process.

• The Government should aim to complete most of the implementation process during this Parliament, which seems feasible. To help meet the implementation timetable delivery issues, such as scheme administration, should be considered at an early stage.

• There will also need to be the right resource, on top of business as usual, to drive the reforms, particularly given the challenging timescale and scope of the reforms, and there should be a robust implementation governance structure to ensure the implementation process is completed effectively.

• Too many risks are held by government and the taxpayer in the present schemes and they produce an unfair distribution of benefits between members. It would not be fair to allow this to continue for decades. Therefore, members of the current defined benefit schemes should be moved to the new schemes.

• In managing the changes there will need to be protections for existing members, such as the protection for past service of the final salary link to future earnings. This means that the final salary link would be maintained for years of service earned in final salary based schemes, up to the date the member is awarded all his or her benefits from that scheme.

• The Commission’s expectation is that existing members who are currently in their 50s should, by and large, experience fairly limited change to the benefit which they would otherwise have expected to accrue by the time they reach their current scheme Normal Pension Age. This would particularly be the case if the final salary link is protected for past service, as the Commission recommends.

148


Implementing change

7.1 Implementation is key to the reform of public service pensions. Reform will only be successful if the planned reforms are capable of being implemented in a timely and efficient manner with a smooth transitional period. This chapter discusses that challenge and some of the areas that will need to be tackled to make sure the reforms can be delivered as planned.

7.2 There are three main stages of the implementation process itself:

• the Government’s consideration of the report, its response and detailed development of proposals;

• detailed scheme by scheme consideration of the proposals followed by formal consultation; and

• legislation and delivery of the reforms.

Developing the details

7.3 It will be up to the Government to decide whether or not to accept the recommendations of this report. But whatever the decisions there will need to be a process to develop the policy from a framework proposal to an implementable blueprint. How this is approached could make a real difference to how and when the reforms are implemented.

7.4 Crucial to this is whether the process is one of control from the centre or whether it is devolved to the different schemes to manage. Both approaches have benefits. A central approach allows the Government to keep tight control over the process, driving it forward along a set timetable; whilst a devolved approach would give schemes the ability to develop the proposals in line with their specific needs. But they also both have downsides. The central process makes it more difficult to build in scheme by scheme variation; whilst the devolved process could result in duplication of work between the different schemes, reducing the efficiency and speed of the process.

7.5 Currently the schemes are run in a highly devolved manner and whilst this can deliver schemes focused on members’ needs it has also led to a wide variety of outcomes and contribution levels for members. When this approach was followed in the last round of reforms it led to a large spread in the timetable for implementation between the different schemes and varying scheme structures. And rather than this approach being actively chosen as best practice or for efficiency, it seems rather to have developed organically. It would be better for the policy development process to be a more structured one, but still allow schemes to enter into individual negotiations and discussions.

7.6 There will also need to be a degree of central control: setting cost ceilings for the schemes; outlining the principles to be adhered to as part of the framework; setting limits

149


to the degree of variation allowed; setting the timescales for consultation and overall implementation; and providing oversight of the scheme-by-scheme implementation process.

7.7 This would have the benefit of allowing standardisation whilst still giving schemes control over the details of the proposals for their particular areas. It would also allow a longer timeframe for implementation, where this might be needed because of the scale of the change or for legal reasons, for example, for schemes such as the armed forces, police and judges.

Agreeing the deal

7.8 Once the Government and schemes have come to an initial view on the detailed future form of the schemes they will need to consult with stakeholders on their proposed approach. Whilst this could be done via one consultation from central government, the Commission would not recommend this: it is likely to be cumbersome given the number of different schemes and the possibility for variation within them.

7.9 The other question is how the consultation process will be carried out. This could either be on a formal basis, such as the normal 12 week consultation period followed by government consideration of the responses, or there could be a more informal initial process, which would take place over a longer period of time and leave more room for open dialogue between the parties, followed by a full statutory consultation. The latter approach would help develop trust and confidence in the process and is favoured by the Commission.

Recommendation: The consultation process itself should be centrally co-ordinated: to set the cost ceilings and timetables for consultation and overall implementation. However, the consultation on details should be conducted scheme by scheme involving employees and their representatives (Recommendation 25).

7.10 For the statutory consultation, as required by law, each consultation document will need to be supported by a full equality impact assessment to allow for a deeper and more considered examination of the issues for particular groups and schemes within a formal framework.

Getting members involved

7.11 The Commission has found it helpful to engage as many interested parties as possible in developing its knowledge base and in forming its opinions. Evidence from stakeholders during the Commission’s roundtable events, as described in Annex D, has allowed a wide range of interested parties to put their views forward, often in areas they had specific expertise in.

150


7.12 Discussions during the roundtables on implementation emphasised the importance of good communication for successful pension reform. The Commission’s view is that there should be early upfront communication of the changes with employees, including using methods that allow direct member participation in the consultation process. This will help members become involved in shaping their scheme and fits with the culture of ownership and responsibility the Commission would like to encourage.

The timetable and achieving it

7.13 These reforms cannot be achieved overnight. As the discussion above shows there are several steps that will need to be taken before the necessary legislative process can be started and the consequent administrative changes made. And these steps are crucial in ensuring that the reforms are a success and deliver sustainable public service pensions schemes.

7.14 The administrative change will be one of the largest challenges. The roundtable events the Commission held had valuable evidence to share on this point. The complexity of, and possibility for, efficiencies in public service pension scheme administration is discussed in Chapter 6, but the importance of taking early account of administration, and communicating with scheme administrators, early in the change process is not to be overlooked. Our stakeholders told us that following this approach should lead to a much smoother implementation process. So, consideration of the delivery process, particularly the administration of the schemes, needs to be built into the policy development process early on.

Recommendation: The Commission’s view is that even allowing for the necessary processes it should be possible to introduce the new schemes before the end of this Parliament and we would encourage the Government to aim for implementation within this timeframe (Recommendation 26).

7.15 This will not be easy, but it should be achievable. But the delivery of this timetable is heavily dependent on having the right resource available to manage all the stages. There is the need to maintain business as usual, making sure pensioners get their correct entitlement on time, at the same time as resourcing the redesign and its implementation. At present there appears to be a concentration of knowledge of public service pensions, their legal basis and their management in a few specialists. A wider spread of knowledge will be needed to drive through this change. And the skills needed for change management are not necessarily those needed for business as usual.

Recommendation: There will need to be the right resource, on top of business as usual, to drive the reforms; particularly given the challenging timescale and scope of the reforms. (Recommendation 27b).

151


7.16 This should be seen in the context of a robust governance structure for the reform programme itself. As discussed in Chapter 6, governance in the public service pension schemes could be improved generally, and it is important that the governance process for implementation is also robust. Chapter 6 highlights the example of the civil service pensions administration transformation programme that has:

• created an independent board with responsibility for managing business as usual within the scheme, its risks and scheme controls, including having oversight of the scheme administrator; and

• also established a separate, robust governance structure that has managed the transformation programme and the risks associated with change.

7.17 The governance arrangements for business as usual and the transformation programme are necessarily different, but both draw on best practice in the private sector.

Recommendation: Best practice governance arrangements should be followed for both business as usual and the transformation process, for each scheme (Recommendation 27a).

7.18 This is important for several reasons. First, it increases transparency in the process. Second, it allows the monitoring of delivery against targets and gives early warning if it looks like the timetable is slipping. Third, it allows individuals responsible for delivering the reforms to be held to account.

7.19 The Commission believes that focusing on communication and administration early, resourcing the change sufficiently and putting a robust governance structure in place will lead to successful implementation.

Transition

7.20 Turning to the terms of the deal to be implemented, the Commission has pointed out in previous chapters that in the present schemes too many risks are held by government and the taxpayer and they produce an unfair distribution of benefits between members.

7.21 As noted above and in the interim report,1 the old, mainly final salary, designs give unfair advantages to those with considerable earnings progression, typically the higher earners, and they disadvantage those with flatter careers, when overall costs are constrained. Moving to a career average structure, as recommended in this report, will produce fairer outcomes.

1 IPSPC (2010) Interim Report, Chapter 5.

152


7.22 And as the interim report showed,2 existing members have benefited from increasing longevity, but that has not been taken into account in the features of current public service pension schemes. Therefore, fundamental reform will need to include increasing pension ages for almost all existing members, as discussed in Chapter 4. Allowing current members to continue to accrue benefits in the present schemes for many decades would be unfair and inequitable to the new members coming behind them.

Recommendation: As soon as practical members of the current defined benefit public service pension schemes should be moved to the new schemes for future service (Recommendation 5a).

7.23 It will also be essential for there to be good planning for, and communication of, the transition to new pension terms. This should ensure that, while implementing revised pension systems satisfactorily in ways that meet the need for fundamental reform indicated in this report and the interim report, members have their rights protected and are given the opportunity to adjust their expectations and retirement plans.

Accrued rights

7.24 The Commission takes as its starting point the principle that accrued rights must be protected, as stated in its terms of reference. For example, service earned on the basis of a specific pension age could not be changed without a member’s consent and therefore pension rights earned up to the date of any change would be based on the current pension ages that apply to that service. To illustrate that, someone who had been earning benefits that would be paid on an unreduced basis from a Normal Pension Age (NPA) of 60 would continue to be able to take those pension rights earned up to the date of the change at age 60.

7.25 However, legally the full extent of those accrued rights is inherently uncertain. For example, general provisions of occupational pensions law require that an active member is at least awarded a deferred pension, but the actual nature of a member’s rights and protections has to be considered and can vary scheme by scheme, depending on scheme rules and how the scheme has been operated.

Deferred and pensioner members

7.26 For those members who have already ceased pensionable service, whether they are deferred or pensioner members, all rights to future benefits, including those potentially payable on death, will be deemed to be accrued rights. Those rights also include the ages

2 IPSPC (2010) Interim Report, Chapter 4.

153


from which members are entitled to take pensions, so the pension will continue to be payable at 60 for most schemes or the relevant scheme age.

Current active members

7.27 For those who are active members of the existing schemes when scheme rules change, it might be argued that accrued rights could be assessed on the basis that the member had opted to leave the scheme immediately before the date of change. Such an approach would be consistent with general legislation concerning accrued rights in private sector schemes,3

7.28 However, the powers to amend the schemes and the extent to which there are any scheme-specific protections for accrued rights must also be considered. That is of central importance, as has already been seen in legal cases relating to accrued rights in both private and public sector schemes.

7.29 Although the powers of amendment for at least two public service schemes explicitly refer to that general legislation on accrued rights,4 most powers do not. But there are some specific provisions and requirements in the amendment powers of some of the other schemes that limit the scope and process for amending benefits.5 Also, more general claims could be made that there are protections for groups of members based on legal provisions and principles that would limit or prevent changes to core features of the benefit design, such as final salary links. Such protections would, for example, be based on legitimate expectations, contractual promises that may have been made, fairness and the principle of not making retrospective changes that reduced entitlements.

Final salary link for past service

7.30 The Commission has also noted in the evidence submitted to it the widespread expectations among public servants that the final salary link would be maintained. The Commission is sympathetic to the argument that it would be in line with the principles underlying accrued rights to maintain the final salary link for past service and would be fair given the scale of the changes involved in moving to the reformed schemes.

7.31 It would also be an important factor in helping to ensure that future pension arrangements were seen as fair to taxpayers and members. This would assist in the successful delivery of the reforms and the maintenance of trust and confidence in public service pensions. This should, in turn, help to maintain levels of scheme membership. It is important to bear in mind that decisions about whether to belong to a pension scheme are individual rather than collective and that scheme membership is not compulsory. The

3 See Section 67 of the Pensions Act 1995, as amended by the Pensions Act 2004.

4 The Armed Forces Pension Scheme 2005 and the Reserve Forces Pension Scheme.

5 For example the Police Pensions Act 1976 and the Judicial Pensions and Retirement Act 1993.

154


Commission wishes to maintain, and ideally increase, the proportion of members who join public service schemes.

Recommendation: The Government should honour in full the pension promises that have been accrued by scheme members: their accrued rights. In doing so, the Commission recommends maintaining the final salary link for past service for current members (Recommendation 4).

7.32 This means that the final salary link would be maintained for years of service earned in final salary based schemes, up to the date the member is awarded all his or her benefits from that scheme, which could be before, at, or after the NPA. In effect that would mean there would be a final salary link as long as the member remained within the existing scheme or its successor.6

7.33 The fair treatment of accrued rights must apply to all relevant benefits and conditions. The Commission is clear that:

• the definition of pensionable pay for the accrued pension should be defined as at the point when the member moves to new schemes for future accrual;

• comparable death and ill-health benefits should be retained;

• movement to another scheme within the public sector transfer club7 should be regarded as continuous service for the purpose of the final salary link;

• those on a career break of up to five years at the point of transfer to new schemes or who take such a break in future should retain the final salary link for past service; and

• those serving public servants who are already in career average pension schemes should receive comparable protections, so, for example, the amounts of pension they had accrued should continue to be indexed in the same way as at present up to the date a member is awarded benefits.

6 That would build on the existing civil service precedent used, for example, where someone with final salary rights enters

the Nuvos career average scheme; and for those members who opted to keep PCSPS Classic 1/80th accrual rights for

past service when they moved in 2002 to PCSPS Premium 1/60ths rights for future service. One consequence would, for

example, be that those NHS scheme members who, as part of the NHS Choice exercise, opted to convert their NPA 60,

1/80ths accrual pension rights into NPA 65, 1/60ths accrual rights, would continue to have those resulting NPA 65 years

of service linked to their final salary and payable unreduced from age 65.

7 A group of 120 defined benefit related occupational pension schemes designed to allow easier movement of staff,

mainly within the public sector, by making sure that employees receive broadly equivalent credits when they transfer their

pensionable service to their new scheme.

155


Effects on members near to retirement

7.34 The Commission’s expectation is that existing members who are currently in their 50s should, by and large, experience fairly limited change to the benefit which they would otherwise have expected to accrue by the time they reach their current scheme NPA. This would particularly be the case if the final salary link is protected for past service, as the Commission recommends. This limitation of impact will also extend to people below age 50, proportionate to the length of time before they reach their NPA. Therefore special protections for members over a certain age should not be necessary. Age discrimination legislation also means that it is not possible in practice to provide protection from change for members who are already above a certain age.

7.35 Those employees who intend to take their pension in the next few years could do so before the new terms are introduced. An employee now aged around 50 with many years of service in a scheme with an NPA of 60 would retain the link to his or her final salary for past service, while accruals from about the age of 55 would be under the new terms with a higher NPA. Although the exact impact of this will depend on individual circumstances and the scheme parameters, it is likely that most people currently in their early 50s will have a slightly lower pension if they choose to retire at their current pension age. Individuals could choose either to retire at the age of 60 with a slightly reduced pension, or work for a little longer in order to obtain the same pension income as that which would previously have been payable at 60.

156


157


8 Conclusions

How the deal meets the Commission’s principles

8.1 The Commission’s view is that if the recommendations outlined in this report were to be implemented as a full reform package it would ensure the future of good quality, defined benefit pension provision within the public services.

8.2 Throughout the report the Commission has assessed the design choices against its principles. Here we show how the recommended package as a whole meets the principles that were first set out in the interim report.1

Affordable and sustainable

8.3 The proposed reforms provide safety valves in the system, which should help to assure sustainability. Linking Normal Pension Age (NPA) to State Pension Age (SPA) will adjust for the improvement in longevity seen over recent past decades and the resulting increase in the proportion of life spent in retirement. It also provides a way to manage any future expected increases in longevity. Having a limit on costs through a fixed cost ceiling, which if exceeded will result in action to get costs back down, will further guarantee affordability of the schemes, providing the taxpayer with confidence that controls are in place.

Adequate and fair

8.4 The Commission has recommended that public service pensions, in conjunction with a full state pension, should provide a level of income that meets agreed adequacy levels for those with a full career. Moving away from final salary schemes and moving current members into the new defined benefit schemes recommended in this report will mean that the schemes are fairer between different types of members. In addition, maintaining defined benefit provision, but sharing key risks, will mean that schemes are fairer between members and taxpayers.

1 IPSPC (2010), Interim Report, Chapter 3.

158


Supports productivity

8.5 Removing the final salary element of scheme design will help to break down barriers to labour mobility between the public and private sectors, as well as reducing some of the barriers to plurality of provision for public services.

Transparent and simple

8.6 A single scheme across the income distribution, with improved minimum standards for communications with scheme members, will help ensure public service pensions are simple and transparent for public service workers. More consistent information published by schemes with oversight of fiscal sustainability from the Office for Budget Responsibility will ensure there is greater transparency for taxpayers too.

Looking forward

8.7 These reforms cannot be achieved overnight and the journey to full implementation may well be a difficult one. But the Commission is clear that it represents the best chance of achieving a public service pensions system that is both sustainable for the future and delivers a good deal for both scheme members and taxpayers.

159


A List of recommendations

A.1 The Table A.1 lists the recommendations which the Commission is making to the Government.

Table A.1: Recommendations

Chapter 1: The case for reform revisited

No. Recommendation Page No.

1

The Government should make clear its assessment of the role of public service pension schemes. Based on its framework of principles, the Commission believes that the primary purpose is to ensure adequate levels of retirement income for public service pensioners.

8

35

2

Pensions will continue to be an important element of remuneration. The Commission recommends that public service employers take greater account of public service pensions when constructing remuneration packages and designing workforce strategies. The Government should make clear in its remits for pay review bodies that they should consider how public service pensions affect total reward when making pay recommendations.

8

36

Chapter 2: The dealNo. Recommendation Page No.

3

The Government should ensure that public service schemes, along with a full state pension, deliver at least adequate levels of income (as defined by the Turner Commission benchmark replacement rates) for scheme members who work full careers in public service. Employers should seek to maximise participation in the schemes where this is appropriate. Adequate incomes and good participation rates are particularly important below median income levels.

9

40

4

The Government must honour in full the pension promises that have been accrued by scheme members: their accrued rights. In doing so, the Commission recommends maintaining the final salary link for past service for current members.

9

156

5

As soon as practical, members of the current defined benefit public service pension schemes should be moved to the new schemes for future service, but the Government should continue to provide a form of defined benefit pension as the core design.

9

44

154

6

All public service pension schemes should regularly publish data which, as far as possible, is produced to common standards and methodologies and is then collated centrally. This information should be of a quality that allows simple comparisons to be made across Government, between schemes and between individual Local Government Pension Scheme (LGPS) Funds.

10

133

160


Chapter 3: The design


7A new career average revalued earnings (CARE) scheme should be adopted for general use in the public service schemes.

10

58

8

Pension benefits should be uprated in line with average earnings during the accrual phase for active scheme members. Post-retirement, pensions in payment should be indexed in line with prices to maintain their purchasing power and adequacy during retirement.

11

66

71

9

A single benefit design should apply across the whole income range. The differing characteristics of higher and lower earners should be addressed through tiered contribution rates. The Government should consider the trade off between affordability and the impact of opt outs on adequacy when setting member contribution levels.

12

65

79

80

10

Members should have greater choice over when to start drawing their pension benefits, so they can choose to retire earlier or later than their Normal Pension Age and their pension would be adjusted accordingly on an actuarially fair basis. Flexible retirement should be encouraged and abatement of pensions in its current form for those who return to work after drawing their pensions should be eliminated. In addition, caps on pension accrual should be removed or significantly lifted.

12

83

Chapter 4: The controls


11

The Government should increase the member’s Normal Pension Age in the new schemes so that it is in line with their State Pension Age. The link between the State Pension Age and Normal Pension Age should be regularly reviewed, to make sure it is still appropriate, with a preference for keeping the two pension ages linked.

13

94

12

The Government, on behalf of the taxpayer, should set out a fixed cost ceiling: the proportion of pensionable pay that they will contribute, on average, to employees’ pensions over the long term. If this is exceeded then there should be a consultation process to bring costs back within the ceiling, with an automatic default change if agreement cannot be reached.

13

97

161


Chapter 5: Applying the design


13

The Commission is not proposing a single public service pension scheme, but over time public service pensions should move towards a common framework for scheme design as set out in this report. However, in some cases, for example, the uniformed services, there may need to be limited adaptations to this framework.

14

110

14

The key design features contained in this report should apply to all public service pension schemes. The exception is in the case of the uniformed services where the Normal Pension Age should be set to reflect the unique characteristics of the work involved. The Government should therefore consider setting a new Normal Pension Age of 60 across the uniformed services, where the Normal Pension Age is currently below this level in these schemes, and keep this under regular review.

14

112

15

The common design features laid out in this report should also apply to the LGPS. However, it remains appropriate for the Government to maintain the different financing arrangements for the LGPS in future, so the LGPS remains funded and the other major schemes remain unfunded.

15

114

115

16It is in principle undesirable for future non-public service workers to have access to public service pension schemes, given the increased long-term risk this places on the Government and taxpayers.

15

119

Chapter 6: A transparent and effective systemNo. Recommendation Page No.

17

Every public service pension scheme (and individual LGPS Fund) should have a properly constituted, trained and competent Pension Board, with member nominees, responsible for meeting good standards of governance including effective and efficient administration. There should also be a pension policy group for each scheme at national level for considering major changes to scheme rules.

16

126

130

18

All public service pension schemes should issue regular benefit statements to active scheme members, at least annually and without being requested and promote the use of information technology for providing information to members and employers.

16

132

19

Governance and the availability and transparency of information would be improved by government establishing a framework that ensures independent oversight of the governance, administration and data transparency of public service pension schemes. Government should consider which body or bodies, including, for example, The Pensions Regulator, is most suitable to undertake this role.

16

135

20

When assessing the long term sustainability of the public finances, the Office for Budget Responsibility should provide a regular published analysis of the long term fiscal impact of the main public service pension schemes (including the funded LGPS).

17

137

162


21

Centrally collated comprehensive data, covering all LGPS Funds, should be published including Fund comparisons, which, for example, clarify and compare key assumptions about investment growth and differences in deficit recovery plans.

17

140

22

Government should set what good standards of administration should consist of in the public service pension schemes based on independent expert advice. The Pensions Regulator might have a role, building on its objective to promote good administration. A benchmarking exercise should then be conducted across all the schemes to assist in the raising of standards where appropriate.

17

141

23

Central and local government should closely monitor the benefits associated with the current co-operative projects within the LGPS, with a view to encouraging the extension of this approach, if appropriate, across all local authorities. Government should also examine closely the potential for the unfunded public service schemes to realise greater efficiencies in the administration of pensions by sharing contracts and combining support services, including considering outsourcing.

17

143

24The Government should introduce primary legislation to adopt a new common UK legal framework for public service schemes.

18

147

Chapter 7: Delivering the change


25

The consultation process itself should be centrally co-ordinated: to set the cost ceilings and timetables for consultation and overall implementation. However, the consultation on details should be conducted scheme by scheme involving employees and their representatives.

18

151

26

The Commission’s view is that even allowing for the necessary processes it should be possible to introduce the new schemes before the end of this Parliament and we would encourage the Government to aim for implementation within this timeframe.

18

152

27

Best practice governance arrangements should be followed for both business as usual and the transformation process, for each scheme. And there will also need to be the right resource, on top of business as usual, to drive the reforms; particularly given the challenging timescale and scope of the reforms.

19

152

153

Source: IPSPC.

163


B Case studies of best practice

B.1 The Commission explored various examples of best practice pension scheme design. This included looking at both the international arena and the private sector. Below are case study examples that the Commission found particularly useful.

Private sector case studies

Box B.1: Tesco

Tesco operates a Career Average Revalued Earnings pension scheme with an accrual rate of 1.5 per cent of salary and a pension age of 65. It is open to all employees. There is no waiting period, but employees aged over 25 are automatically enrolled after a year and can choose to opt-in in their first year if they so wish. Member contributions are 5 per cent of salary and employer contributions are currently just over 11 per cent. Pensions are indexed by inflation for all members subject to a cap of 5 per cent during the accrual phase and also post retirement. Death in service benefits include a spouse’s pension for qualifying partners (60 per cent of pension) and a lump sum of three times pay.

Box B.2: The Co-operative Group

The Co-operative Group operates a Career Average Revalued Earnings (CARE) pension scheme called PACE (Pension Average Career Earnings Scheme). It has an accrual rate based on 1/60th of pensionable earnings for every year that a member remains in the scheme, which is uprated in line with inflation for both active and deferred members, subject to a five per cent cap. The scheme has a pension age of 65 and is open to all eligible employees of the Co-operative Group and its subsidiaries, with a waiting period of three months.

Member contributions are 6 per cent of pensionable earnings. The employer pays the balance of cost of providing all members’ benefits. Pensions in payment are indexed according to inflation subject to a cap of 2.5 per cent. Death in service benefits include a spouse’s pension for qualifying partners (50 per cent of accrued pension plus 25 per cent of notional pension to age 65) and a lump sum of three times pay.

164


Box B.3: John Lewis

In October 2008 the company made some changes to their non-contributory final salary scheme. These changes reduced the waiting time for the scheme to three years and established a nursery defined contribution scheme with matched contributions of up to six per cent.

The organisation also introduced a Life Expectancy Adjustment Factor (LEAF) as a way of sharing pre-retirement longevity risk with members. The risk associated with longevity projection changes between October 2008 and retirement are passed onto the member while any further changes in longevity projections post-retirement are borne by the scheme.

The LEAF factor is calculated each year by the scheme actuary who takes into account life expectancy, pension increases and the interest the pension fund earns on its investments to work out the cost of providing £1,000 per year of pension for a 65 year old. LEAF is then calculated as the ratio between the cost of a £1,000-per-year pension for a 65 year-old scheme member in 2008 (when LEAF was introduced) and the cost of a £1,000-per-year pension for a 65 year-old scheme member in the year of calculation. This figure is then used to calculate pension benefits by multiplying the original expected benefits by the LEAF.

The LEAF is calculated and announced a year ahead and the current LEAF is used in pension statements with the provision that this is an estimate and subject to change. It is also phased in over the year. The maximum benefit change year on year is +/- 0.5 per cent.

Death in service benefits include a spouse’s pension for qualifying partners (50 per cent of a spouse’s prospective pension) and a lump sum of three times pay.

All changes were agreed by employees and their representatives before being implemented.

165


International case studies

Box B.4: Sweden

Why was the Swedish pensions system reformed?

In 2003 Sweden adopted an unfunded Notional Defined Contribution (NDC) system affecting all who work in the country (similar to the UK’s State and Second State Pension). The old system was a defined benefit career average system, based on an individual’s 15 years of highest earnings. The principle reasons for reform were increasing pensions costs while economic growth remained low, as well as the belief that the old system was less fair to people who had lower income growth and worked over a long period of time.

How does the new scheme work?

In the Notional Defined Contribution (NDC) scheme total contributions of 16 per cent (7.5 per cent employee with the rest made up of employer and tax efficiencies) of salary are credited to an individual’s account, up to a capped amount. Individual account balances grow with annual contributions and are then recalculated every year based on per capita wage growth. However, an ‘automatic adjustment mechanism’ exists which responds to changes in the economy, e.g. accrual rates can alter from average wage growth if financial imbalances occur due, for example, to excessive changes in longevity.

What benefits does the new scheme bring?

• Sustainability: benefits are driven by past contributions, limiting taxpayer liability.

• Risk reduction: annuities are calculated on life expectancy at retirement so longevity risk lies with employees up to the point of retirement.

• Immediate response to economic shocks: the rate of return used to grow individuals’ pension pots is linked to economic factors that allow for automatic reductions in benefits during times of poor economic growth.

• Fairness between generations: defined contribution schemes by design ensure that intergenerational transfer is avoided.

• Removing labour market distortions: pensions reflect an individual’s entire career earnings.

166


Box B.5: The Netherlands

What is the Dutch Collective Defined Contribution (CDC) model?

The Dutch public sector has adopted this model (the ‘ABP’) and several companies within the private sector have implemented similar systems. The scheme runs like a defined benefit scheme. It is based on a career-average model; the difference lies in the onus of risk placed onto the employer. This is managed through:

• solvency margins: 130 per cent of the liabilities are funded, meaning contributions are higher, yet in the instance that the market does badly, a buffer fund exists that will absorb the damage and allow liabilities to be paid, and monies are invested in relatively low-risk areas; and

• conditional indexation: every few years, the indexing of the pension funds is re-assessed to counteract the losses/gains to the buffer fund.

There is no explicit risk-sharing between generations; if necessary a proportion is taken from each generation’s ‘pot’ to cover any liabilities.

What benefit does the scheme bring?

• Conditional indexation: indexation alters according to the achievement of the fund. However, during the recession, funding ratios fell drastically for the vast majority of the Dutch CDC schemes’ pension funds, which had to submit recovery plans to their regulator, specifying how their underfunding would be eliminated.a The recovery plans have in almost all cases included suspending indexation of acquired pension rights through the conditional indexation mechanism until the appropriate funding levels are regained.

• Solvency margins: 130 per cent of liabilities are funded to deal with market shocks.

• Employees do not have the burden of managing individual accounts.

• Significantly reduced investment fees and other costs.


a) Høj, J. (2011), ‘Making the Dutch Pension System Less Vulnerable to Financial Crises’, OECD Economics Department

Working Papers, No. 832, OECD Publishing. doi: 10.1787/5kgkdgg5fxd3-en.

167


Box B.6: United States of America

Why was the US civilian employee pensions system reformed?

The Federal Employees Retirement System (FERS) was implemented in 1987 to replace the old defined benefit plan for all new federal civilian employees. FERS consists of three parts: state social security (much like the UK basic State Pension); an unfunded defined benefit plan; and a funded defined contribution plan.

The reform was instigated because the old system did not involve contributing to Social Security, but additional Social Security contributions were needed in order for the Social Security system to remain solvent.


Under the defined benefit element, the pension is based on the highest three year average of service earnings and accrues at a rate of 1 per cent per annum, or 1.1 per cent from age 62 with 20 or more years’ service, to a maximum total accrual of 40-44 per cent of pay after 40 years service. The employee contribution to the defined benefit plan is 0.8 per cent of salary.

Under the defined contribution element (the ‘Thrift Savings Plan’) the Government automatically contributes an amount equal to 1 per cent of salary for each employee. In addition, employees may contribute up to 10 per cent of their salaries and receive government-matching contributions on the first 5 per cent.


• Certainty and risk-redistribution: the defined benefit core grants post-retirement income certainty for employees, and the defined contribution top-up redistributes some of the risk away from the employer.

• Lower costs: the defined contribution element of the new scheme has lowered the costs of the new scheme compared to the old scheme.

• A ‘loans’ feature: the defined contribution element has a ‘loans’ feature in which members can access their retirement savings, whilst still working, on a loan basis.

• Removing labour market distortions: the defined-contribution element allows flexibility between labour markets.

168


Box B.7: Australia

Why was the Australian government employee’s pension system reformed?

The Australian Government introduced the Public Sector Superannuation Accumulation Plan (PSSap) as a fully funded accumulation scheme for most new Australian Government employees that joined from 1 July 2005. The old Public Sector Superannuation Scheme (PSS), which was an unfunded defined benefit scheme, was closed to new members from 30 June 2005. Australia chose to close most public defined benefit superannuation schemes to new members and replace them with fully funded accumulation schemes in response to fiscal challenges.


The employer contributes at a rate of 15.4 per cent of pensionable salary. Members have the option of making voluntary personal contributions, including salary sacrifice contributions. These amounts are paid into the PSSap Fund, in which members can make their own decisions as to how contributions are invested. The PSSap benefit is a lump sum.


• Employee choice: employees can choose how their accounts are invested.

• Lower risks to the taxpayer: the move from a defined benefit to a defined contribution scheme moves more of the risk away from the employer.

• Fairness between generations: each individual has their own personal pensions pot which they are responsible for, removing risk sharing between generations.

• Removing labour market distortions: the system allows flexibility between labour markets.

169


Box B.8: Poland

Why was the Polish pensions system reformed?

The defined benefit pensions system that Poland inherited from the Communist era was not suited to a market economy. Problems included a large number of pensioners because coverage was high and pensionable age low, and pension benefits varying hugely, bearing little resemblance to contributions or need.


The Polish reform was launched in 1999 and consisted of: a notional defined contribution, unfunded first pillar (equivalent to the UK basic State Pension); a mandatory defined contribution, privately-managed, funded second pillar (which plays a similar role to the UK State Second Pension); and voluntary employee pension plans in the third pillar.

Contributions to the reformed pension system account for 19.52 per cent of employees’ taxable income, with employers and employees each paying half. Of that amount, 12.22 per cent goes into the public notional account scheme and 7.3 per cent is credited to the defined contribution pension plan. Contributions to the unfunded first pillar are indexed in line with 75 per cent of the quarterly growth of the covered wage bill.


• Diversification of retirement savings: labour market developments determine the notional rate of return in the unfunded scheme and financial market developments determine rate of return in the funded scheme.

• Transparency of costs: early retirement options can be maintained, but workers must pay a higher contribution rate to reflect the additional cost.

• Ease the transition to a funded system: the long-term aim with the new scheme is to have 50 per cent of contributions going to the unfunded pillar and 50 per cent to the funded pillar (initially 62.5 per cent and 37.5 per cent respectively).

• Fairness between generations: defined contribution schemes by design ensure that intergenerational transfer is avoided.


170


171


C Technical modelling

PPI analysis of pension scheme structure

C.1 The Commission asked the Pensions Policy Institute (PPI) to carry out an analysis of different scheme structures that could be used as future public service pension schemes. Starting from a proxy final salary scheme (intended to broadly replicate a typical public sector pension scheme), the aim was to define the parameters for a set of alternative scheme structures in order that they would, on average, provide the same value of benefits to the scheme membership. The distributional impact of the alternative structures on members of different age, income level and gender was then analysed.

C.2 The analysis was based on the following three alternative scheme structures:

• career average scheme;

• career average scheme with a cap on pensionable earnings at £75,000; and

• career average scheme capped at £35,000 and a DC top-up scheme for earnings above the cap with employee and employer contributions of 5 per cent and 10 per cent respectively.

C.3 Further to the over-arching structure, different options within these broad structures were considered. This was primarily around the level of pre-retirement indexation in the CARE schemes where a prices measure and an earnings measure were considered.

C.4 From these results it is also possible to infer the potential implications of a cash balance scheme structure with a combination of accrual rate and fixed conversion factor that is equivalent to the accrual rate in the CARE scheme. However, it should be noted that not all cash balance schemes have a fixed conversion factor.

C.5 The parameters included here and elsewhere in the final report are not the Commission’s recommendations or suggestions for those that should be used in future public service pension schemes. These will need to be set by the Government, in consultation with scheme members, after full actuarial calculations have been carried out.

Methodology

C.6 The PPI Proxy Public Sector Scheme is a final salary scheme which broadly emulates the current arrangements offered to new entrants of the NHS, Teachers and Local Government

172


pension schemes. It is used as a standard benchmark against which to measure the broad impacts of reforms. However it should be recognised that the proxy scheme is not identical to any of the current public service pension schemes, for example, many public service scheme members have a Normal Pension Age of 60.

Table C.1: PPI proxy scheme structure

PPI proxy scheme Parameter

Scheme design Final salary

Normal pension age 65

Contributions Tiered by income

Accrual 1/60th of salary for each year of service

Indexation Final salary link for active members, CPI indexation for deferred and pensioner members

Source: PPI modelling.

C.7 Tiered contributions within the proxy scheme vary as shown in Table C.2, set to fall broadly within the existing tiered contributions of the NHS and the Local Government pension schemes.

Table C.2: Tiered contributions in the PPI proxy scheme

Band Earnings range (£) Rate (%)

1 0-20,000 5.25

2 20,000-40,000 6.5

3 40,000-70,000 7.0

4 70,000-100,000 7.5

5 Over 100,000 8.0


C.8 The measure adopted to assess the value of benefits to the scheme membership was the effective employee benefit rate (EEBR).1 This represents the amount that would need to be set aside to ‘buy’ one year’s accrual of benefits under a particular set of assumptions, if the scheme were funded, expressed as a percentage of pay. Member contributions are deducted, therefore the EEBR represents the notional amount that is contributed by the employer on behalf of the member. The calculation takes account of the main features of the schemes’ designs, including their Normal Pension Age (NPA), accrual rate, indexation rate and ancillary benefits.

1 More information on the broad methodology of the effective employee benefit rate can be found in previous PPI

publications, for example, in appendix 1 of An assessment of the Government’s reforms to public sector pension, PPI,

October 2008.

173


C.9 The first stage of the analysis determined the accrual rate required for the alternative scheme structures such that the resulting value of benefits to the scheme membership was, on average, the same as that under the proxy final salary scheme. This was achieved by varying the accrual rate until the overall EEBR (the weighted average of each member’s EEBR across the whole income, age and gender distribution) in the alternative scheme was the same as the overall EEBR in the proxy final salary scheme.

C.10 Assuming that employer contributions were set using broadly the same assumptions2 adopted when considering value to members, and that the salary and age distribution structure of the schemes were to remain unchanged, then this would also mean that the long-term cost of the schemes in terms of employer contributions would be broadly similar. This does not mean that benefit expenditure cash flows in respect of pensions paid from the scheme would be equivalent in each future year, as these would likely differ. However it does mean that the expected present value of those cash flows would be broadly similar.

C.11 While the overall EEBRs are equivalent in the proxy final salary and alternative schemes, the differing structures and parameters of the schemes lead to different distribution of the benefits between individual members in those schemes (due to differences in age, gender and income). The second stage determined how benefits were distributed between members, by calculating EEBRs for differing income deciles, age bands, career progression and genders.

C.12 The third stage analysed the benefits received from the scheme relative to the member contributions made, for differing career progression. This was done to assess whether the benefit derived from the scheme for different types of member was proportional to the member contributions made.

Assumptions

C.13 There are a number of key financial assumptions in the calculation of the effective employee benefit rate, these are in Table C.3.

2 The accrual rate calculated to be required to maintain a broadly equivalent value to members between the different

schemes considered is not sensitive to the choice of discount rate. However, the actual level of contributions required to

meet the long-term cost of a pension scheme is sensitive to the discount rate used, and HM Treasury has been consulting

on the appropriate discount rate to use in setting employer pension contributions.

174


Table C.3: Key financial assumptions

Assumption Nominal (%) Real (% above RPI)

Consumer price index (CPI) 2.00

Retail price index (RPI) 2.75

Earnings inflation 4.29 1.50

Discount rate 5.32a 2.50


a) This is set to be consistent with previous PPI analysis. The discount rate will affect the EEBR but will not affect the

relative differences between the different types of scheme.

C.14 The discount rate used is in line with previous PPI and IPSPC EEBR estimates. The calculated accrual rates for the CARE schemes are not sensitive to the choice of discount rate and discussion of the effects of the structure of the pension scheme therefore remain unchanged by the level of the discount rate. However, the discount rate would affect the levels of the EEBR.

C.15 Life expectancies are assumed to be in line with the Office for National Statistics 2006 based principal projections.

C.16 Promotional salary scale assumptions are used to allow for pay rises above general salary inflation, for example, pay increases as a result of promotion. Table C.4 sets out an index of expected promotional increases by age.

Table C.4: Salary scale assumptions

Age 20 25 30 35 40 45 50 55 60 65

Male 100 131 168 201 224 237 249 253 256 256

Female 100 131 157 176 187 192 196 200 202 202

Source: GAD Pay-As-You-Go Public Service Pension Schemes December 2009 Cashflow Projections Methodology, data and assumptions.

Results

Specification of scheme designs

C.17 The first stage of the analysis was to derive a set of CARE scheme structures that would be of approximately equivalent value to the proxy scheme. The results of this are summarised in Table C.5. The IPSPC specified all scheme parameters (employee contribution rates, Normal Pension Age (NPA), indexation, levels of cap and DC top-up) except for the accrual rate. The PPI then derived the accrual rate that would give an overall EEBR equal to that of

175


the proxy final salary scheme when applied to the membership profile of the public service schemes. These accrual rates are given in the final column.

Table C.5: Scheme designs

Band Type NPA Pre-retirement indexationa

Capb DC top up Accrual rate

Proxy Final salary 65 Final salary / CPI None - 1 / 60ths

Scheme 1a CARE SPAc CPI None - 1 / 40ths

Scheme 2a CARE SPA CPI £75,000 No 1 / 40ths

Scheme 3a CARE SPA CPI £35,000 Yesd 1 / 39ths

Scheme 1b CARE SPA Earnings None - 1 / 61sts

Scheme 2b CARE SPA Earnings £75,000 No 1 / 61sts

Scheme 3b CARE SPA Earnings £35,000 Yes 1 / 59ths


Note: Employee contributions in all scenarios are based on the proxy tiered contribution structure.

a) The proxy scheme has a final salary link for active members and CPI indexation for deferred and pensioner members.

The CARE schemes have the same level of pre-retirement indexation for both active and deferred members. For all

schemes post-retirement indexation is assumed to be in line with CPI.

b) Cap on pensionable earnings that qualify for CARE benefit.

c) The NPA is assumed to move in line with the changes in male State Pension Age (SPA) legislated in the Pensions Act

2007, increasing from age 65 to age 66 by 2026, to age 67 by 2036 and to age 68 by 2046.

d) DC top-up based on a total contribution of 15 per cent of salary (10 per cent employer, 5 per cent employee) into a

defined contribution arrangement.

C.18 This initial analysis led to the following conclusions:

• The CARE scheme structures that would be approximately equivalent in overall value to the PPI proxy final salary scheme would be:

• accrual rate of 1/40ths, CPI indexation pre-retirement, NPA in line with SPA; and

• accrual rate of 1/61sts, earnings indexation pre-retirement, NPA in line with SPA.

176


• the second scenario here has a slightly lower accrual rate than the proxy scheme due to deferred benefit indexation being earnings linked, while in the proxy scheme it is assumed to be CPI-linked. Effectively the savings from the removal of the final salary link have been recycled into providing better benefits for those who do not work a full career in public service;

• applying a cap on pensionable pay at £75,000 or above would have a negligible effect on the accrual rate that would then be able to be offered on earnings below the cap for the same average value to the employee. This is because very few public sector employees earn above this level; and

• applying a cap at a lower level of £35,000 with a DC top-up above this level would mean a slightly more generous accrual rate could be offered on earnings below the cap for the same average value to the employee.

Effective employee benefit rates by age and income

C.19 The analysis includes EEBRs broken down into age bands and income deciles, which allows the consideration of a CARE benefit structure on different categories of member. Chart C.1 shows how the EEBR varies with age for members with median earnings.

Chart C.1: Age effects of schemes based on median earner

0%

5%

10%

15%

20%

25%

30%

35%

Under 25 26-35 36-45 46-55 56 plusEffe

ctiv

e E

mpl

oyee

Ben

efit

Rat

e

Age group

Proxy CARE (CPI)CARE (earnings) CARE (CPI) + £35k cap with DC top upCARE (earnings) + £35k cap with DC top up

Source: IPSPC analysis of PPI results.

177


C.20 The EEBR generally increases with age in the proxy scheme. The only exception is for the under 25 age group where the EEBR is higher than for the 26-35 group as a result of the under 25s being assumed to have a high initial salary growth.

C.21 The CPI indexation schemes show very strong age effects. The EEBR increases substantially with age, with the value of benefits significantly higher for older members than younger members. The cap has very little impact because a median-earning employee does not earn at a level that breaches the cap, but they do benefit from the slightly higher rate of accrual that having a cap on high earners allows.

C.22 In the CARE scheme with pre-retirement indexation linked to growth in average earnings, the age effects are less than in the CPI indexation scheme but are still apparent, to a similar extent as in the final salary structure. As with the CPI indexation scheme the capped version is slightly more generous to median earners

C.23 The picture for low earners is very similar to that for median earners. Chart C.2 shows how the EEBR varies with age for members with high earnings (at the 90th percentile of earnings across public service pension scheme members).

Chart C.2: Age effects of schemes based on 90th percentile earner

0%

5%

10%

15%

20%

25%

30%

35%

Under 25 26-35 36-45 46-55 56 plus

Effe

ctiv

e E

mpl

oyee

Ben

efit

Rat

e

Age group

Proxy CARE (CPI)CARE (earnings) CARE (CPI) + £35k cap with DC top upCARE (earnings) + £35k cap with DC top up


C.24 For high earners the uncapped schemes show similar patterns of EEBR as for median earners. However the impacts of capping are greater for higher earners:

178


• in the CPI indexation schemes, younger members have a higher EEBR from capped schemes. But older members do better in uncapped schemes. This is because for younger members, where the value of the CARE scheme is low, the value from the DC top-up portion of the capped scheme is more valuable than the CARE portion. For older members, the CARE portion is more valuable than the DC top-up portion; and

• in the earnings indexation schemes, higher earning members of all ages are expected to fare better in the pure CARE scheme than the hybrid CARE and DC top-up scheme. This is because the DC top-up portion is less valuable to members of all ages than the CARE portion of the scheme.

Effective employee benefit rates by gender

C.25 Chart C.3 shows EEBRs by gender for the different scheme structures.

Chart C.3: Effective employee benefit rates by gender

0%2%4%6%8%

10%12%14%16%18%20%

Proxy CARE (CPI) CARE (earnings) CARE (CPI) + £35k cap with

DC top up

CARE (earnings) + £35k cap with

DC top up

Effe

ctiv

e E

mpl

oyee

Ben

efit

Rat

e

Male Female

Source: IPSPC analysis of PPI results..

C.26 Under the set of particular assumptions used to derive the figures the following conclusions could be drawn:

• men fare better under final salary structures, primarily because men are expected to have stronger salary progression than women. This more than offsets the assumption that women will live for longer; and

179


• women fare better under CARE structures, because the salary effects are largely stripped out, leaving the effects of the assumption that women will live longer.

Effective employee benefit rates by career progression

C.27 The EEBRs have been determined for three career progression scenarios representing low, mid-level and high flyers:

• low flyers are assumed to receive earnings growth in line with general wage inflation;

• mid-level workers are assumed to experience standard assumed promotional salary scale and earnings inflation; and

• high flyers are assumed to receive standard assumed promotional salary scale increases and earnings inflation plus an additional 1 per cent per annum.

C.28 Chart C.4 shows how members with these career progressions could fare under the different scheme structures. The EEBRs are taken as an average across the whole age distribution but are assumed to have initial earnings of around £16,000.3

Chart C.4: Effective employee benefit rates by career progression

0%

5%

10%

15%

20%

25%

30%

Proxy CARE (CPI) CARE (earnings) CARE (CPI) + £35k cap with

DC top up


DC top up

Effe

ctiv

e em

ploy

ee b

enef

it ra

te

Low flyer Mid flyer High flyer


3 This is at the 30th percentile level and explains why the capped schemes show an overall slightly higher level of EEBR

than the uncapped CARE schemes – see Chart C.1 for further details.

180


C.29 Chart C.4 illustrates how high flyers fare better under final salary structures than mid-level and low flyers. In a CARE scheme the final salary link is broken, and so career progression has a much reduced effect on the EEBR.

Benefits received relative to contributions paid

C.30 Chart C.5 illustrates the amount of benefit received from the scheme as a multiple of the member contributions paid, split by career progression.

Chart C.5: Ratio of benefits received to contributions paid, split by career progression

0

1

2

3

4

5

6

7

Proxy CARE (CPI) CARE (earnings) CARE (CPI) + £35k cap with DC

top up


DC top up

Ratio

Low flyer Mid flyer High flyer


Note: Ratio of benefits received to contributions paid for a 35 year old man initially earning around £16,000.

C.31 These results show that high flyers benefit considerably more than low or mid flyers in a final salary scheme. In a CARE scheme, where there is no link between benefits and final salary, members who experience different career progression receive very similar levels of benefit for pension accrued in a given year. The chart shows that a high flyer could see a return on their contributions of around 50 per cent more than a low flyer under the final salary structure, with no such effects in any of the CARE structures.

C.32 As shown in Chart C.1 and Chart C.2, the earnings indexation scheme is more generous than the CPI indexation scheme for younger people compared to older people. Chart C.5 is for a 35 year old, which is why the earnings scenario appears more generous for all career progression scenarios. If an older member had been chosen (someone aged 55, for

181


example) then the CPI indexation scheme would appear more generous. On average across the membership, the schemes are of broadly equivalent value.

Pensim2 adequacy analysis

C.33 Using data from the Department for Work and Pensions’ Pensim2 model, the Commission applied the parameters calculated by the Pensions Policy Institute to analyse the impact of different scheme designs on adequacy of retirement incomes.

C.34 Pensim2 is a dynamic microsimulation model, which aims to estimate the future distribution of incomes of pensioners. It is thus valuable in assessing the distributional impact of policy changes. It incorporates several data sources to build up synthetic life histories for many thousands of individuals (currently including 60,000, with between 500 and 700 ‘born’ each year). These data sources include the Lifetime Labour Market Database (one per cent survey of National Insurance and PAYE administrative data), the Family Resources Survey, and the British Household Panel Study.

C.35 The Institute for Fiscal Studies assessed Pensim2 comprehensively in 2004; it concluded that “In the main, our assessment of Pensim2 is very positive.”4 Since then, there have been several further incremental changes to the model, including State Pension reforms, private pension reforms and updated assumptions.

C.36 There are limitations to the Pensim2 model. In particular, it is impossible to predict with certainty how work patterns and state pension provision will evolve over the next ninety years. If, for instance, state pensions were to become significantly more or less generous, the likelihood of achieving adequate retirement incomes could be transformed. But the model should give a good picture of how different scheme designs are likely to affect the distribution of pensions.

C.37 The analysis was carried out on a subset of Pensim2, comprising the 933 individuals born between 1991 and 2000 who work at some stage of their careers in the public sector. 856 of these reach State Pension Age (SPA), and the model tracks their life histories until 2100.

C.38 Chapter 2 outlines the Commission’s approach to adequacy, concluding that the benchmark replacement rates set out by Lord Turner’s Pensions Commission provide a minimum standard for retirement incomes (Table C.6). So, for instance, someone earning £30,000 at retirement should have an income in retirement of at least £18,000 (a replacement rate of 60 per cent). In line with the interim report, the bands have been increased with the Average Earnings Index (such that, for instance, the lowest band division is currently at around £11,000).5

4 An Assessment of Pensim2, IFS Working Paper 04/21, Emmerson, Reed and Shephard, 2004.

5 It is assumed that average earnings increase by 2.93 per cent per annum in the long term.

182


Table C.6: Pensions Commission benchmark replacement rates

Gross income Gross income (approximate 2011 terms)

Benchmark gross replacement rate (%)

Less than £9,500 Less than £11,000 80

£9,500 - £17,499 £11,000 - £20,499 70

£17,500 - £24,999 £20,500 - £29,499 67

£25,000 - £49,999 £29,500 - £58,999 60

£50,000 and above £59,000 and above 50

Source: Pensions Commission (2004), Pensions: Challenges and Choices – The First Report of the Pensions Commission.

C.39 The replacement rates are calculated from the point that an individual reaches the State Pension Age (68 for those born between 1991 and 2000),6 and are based on the individual’s final full-time equivalent salary.

C.40 Using the statistical software package Stata, the Commission calculated total pension income for individuals at the age of 68. This was done on the basis that public service employment provided pension accruals from three of the scheme designs assessed by the Pensions Policy Institute (a final salary proxy scheme, uncapped CARE with indexation by average earnings and uncapped CARE with indexation by inflation).7 Since most of the individuals in the Pensim2 database also work in the private sector at some stage in their careers, they usually have other income from employer-sponsored pension schemes (including NEST). Pension income also includes the State Pension, private pensions and any inherited pensions (for instance from a deceased spouse). As the Pensim2 database includes its own projections for public service pension income, this income was subtracted to give the final figures for total pension income based on each of the three different scheme designs.

C.41 The calculations of total pension income do not take into account any means-tested benefit income (such as Pension Credit and housing benefit). This is because the Commission does not believe public service pensions would be adequate if pensioners were to be reliant upon such benefits. Rerunning the analysis including these means-tested benefits increases the proportion achieving their benchmark replacement rates, but it does not have a dramatic impact.

C.42 The analysis assumes that members do not take any lump sum and that they begin to draw their public service pensions at the State Pension Age (SPA). If they took a lump sum, or draw their pension before SPA, this would reduce the proportion of members receiving adequate income levels in retirement.

6 In this database, many people stop work before this age; the median retirement age is 65.

7 Because final salary pension schemes are typically based on the highest salary in the years before retirement, this

analysis uses the individual’s maximum full-time equivalent salary in public service as the basis for his or her final salary

pension payments.

183


C.43 The three different scheme designs on average provide similar levels of pension payments at retirement. However, according to Pensim2, the CARE scheme with earnings indexation provides somewhat higher benefits. This is partly because Pensim2 and the PPI make different assumptions about the age structure of public sector workers, with the PPI assuming that public sector workers are on average older. These different assumptions would tend to skew adequacy calculations; the scheme with earnings indexation would seem to be more likely to produce adequate retirement incomes because its average payments are higher. To equalise average pension payments across the schemes, pension payments on retirement from the scheme with earnings indexation were multiplied by just over 90 per cent.8

C.44 The Commission then used these slightly revised scheme designs to calculate whether total pension income at the State Pension Age is greater than or less than the benchmark replacement rate for each individual. Because relatively few people are in the highest income band, the results in Chapter 3 and below show the top two income bands together.

C.45 Chart C.6 shows the proportion expected to achieve the benchmark replacement rate split by pre-retirement income and scheme design. Under each design around two thirds of members are expected to do so. The Commission’s view is that public service pension schemes should deliver adequate levels of income in retirement for people who spend a full career in the public services, so these results are not surprising.

Chart C.6: Proportion expected to achieve adequacy targets from different scheme designs

0%10%20%30%40%50%60%70%80%90%

100%

Less than £9,500 £9,500-£17,499 £17,500-£24,999 £25,000 and abovePerc

enta

ge a

chie

ving

ade

quac

y ta

rget


Final salary CARE with earnings indexation CARE with inflation indexation


8 Payments under the scheme with earnings indexation were multiplied by 90.35 per cent, those under the final

salary proxy scheme by 99.17 per cent. This reduces average scheme payments relative to those expected from the PPI

modelling, so tends to underestimate the proportions achieving adequacy targets.

Table C.6: Pensions Commission benchmark replacement rates

Gross income Gross income (approximate 2011 terms)

Benchmark gross replacement rate (%)

Less than £9,500 Less than £11,000 80

£9,500 - £17,499 £11,000 - £20,499 70

£17,500 - £24,999 £20,500 - £29,499 67

£25,000 - £49,999 £29,500 - £58,999 60

£50,000 and above £59,000 and above 50

Source: Pensions Commission (2004), Pensions: Challenges and Choices – The First Report of the Pensions Commission.

C.39 The replacement rates are calculated from the point that an individual reaches the State Pension Age (68 for those born between 1991 and 2000),6 and are based on the individual’s final full-time equivalent salary.

C.40 Using the statistical software package Stata, the Commission calculated total pension income for individuals at the age of 68. This was done on the basis that public service employment provided pension accruals from three of the scheme designs assessed by the Pensions Policy Institute (a final salary proxy scheme, uncapped CARE with indexation by average earnings and uncapped CARE with indexation by inflation).7 Since most of the individuals in the Pensim2 database also work in the private sector at some stage in their careers, they usually have other income from employer-sponsored pension schemes (including NEST). Pension income also includes the State Pension, private pensions and any inherited pensions (for instance from a deceased spouse). As the Pensim2 database includes its own projections for public service pension income, this income was subtracted to give the final figures for total pension income based on each of the three different scheme designs.

C.41 The calculations of total pension income do not take into account any means-tested benefit income (such as Pension Credit and housing benefit). This is because the Commission does not believe public service pensions would be adequate if pensioners were to be reliant upon such benefits. Rerunning the analysis including these means-tested benefits increases the proportion achieving their benchmark replacement rates, but it does not have a dramatic impact.

C.42 The analysis assumes that members do not take any lump sum and that they begin to draw their public service pensions at the State Pension Age (SPA). If they took a lump sum, or draw their pension before SPA, this would reduce the proportion of members receiving adequate income levels in retirement.

6 In this database, many people stop work before this age; the median retirement age is 65.

7 Because final salary pension schemes are typically based on the highest salary in the years before retirement, this

analysis uses the individual’s maximum full-time equivalent salary in public service as the basis for his or her final salary

pension payments.

184


C.46 Chart C.7 looks only at those with at least 20 years of public sector employment, showing that overall almost 95 per cent of these individuals are expected to meet the level of income that the Turner Commission assessed was a minimum level in their report. These results are before any lump sum has been taken. If scheme members choose to take a lump sum it is likely they will fall below these minimum levels. But this is less likely to occur when members work full careers in public service, which is likely to be in excess of forty years in the future schemes.

Chart C.7: Proportion expected to achieve adequacy targets with at least 20 years’ public sector employment

0%10%20%30%40%50%60%70%80%90%

100%

Less than £9,500 £9,500-£17,499 £17,500-£24,999 £25,000 and abovePerc

enta

ge a

chie

ving

ade

quac

y ta

rget


Final salary CARE with earnings indexation CARE with inflation indexation


C.47 The Commission has made clear that the benchmark replacement rates should be seen as minimum standards, so it is important to investigate the extent to which different pension scheme designs enable overachievement of adequacy targets. Chart C.8 shows that the average individual receives about 20 per cent more than his benchmark replacement rate from each of the scheme designs. The overachievement is greater for those with longer careers in the public sector. The average public sector employee with at least 20 years’ service receives total pension income at State Pension Age that is around 50 per cent higher than the adequacy target.

185


Chart C.8: Median ratios of pension income to adequacy target from different scheme designs

100%

105%

110%

115%

120%

125%

Final salary CARE with earnings indexation

CARE with inflation indexation

Med

ian

ratio

of p

ensi

on in

com

e to

ad

equa

cy ta

rget


186


187


D Summary of evidence

D.1 This annex details some of the ways in which the Commission has collected information and evidence from a wide range of interested parties since the interim report was published in October 2010.

Second Call for Evidence

D.2 On 1 November 2010, Lord Hutton issued a second call for evidence to inform the Commission’s final report. The call for evidence asked specific questions that were categorised into themes.

D.3 The Commission received a total of 185 submissions as detailed in Chart D.1.1

Chart D.1: Number of submissions received by group

010203040506070

Loca

l Gov

ernm

ent C

ounc

ils

and

Sche

me

Adm

inist

rato

rs

Em

ploy

ee R

epre

sent

ativ

es

Indi

vidu

als

Prof

essi

onal

Org

anisa

tions

an

d C

ompa

nies

Pens

ion

Sche

me

Adm

inist

rato

rs

Rep

rese

ntat

ive

Bod

ies:

Scot

land

Gov

ernm

ent D

epar

tmen

ts

and

Arm

s Le

ngth

Bod

ies

Em

ploy

er R

epre

sent

ativ

es

Rep

rese

ntat

ive

Bod

ies:

Nor

ther

n Ir

elan

d

Aca

dem

ic I

nstit

utio

ns

Loca

l Gov

ernm

ent:

Fire

&

Res

cue

Col

lege

s

Loca

l Gov

ernm

ent:

Out

sour

cing

Com

pani

es

Rep

rese

ntat

ive

Bod

ies:

Wal

es

Num

ber

of r

espo

nses

Stakeholder groups

Source: Responses from ‘Call for Evidence’, November 2010.

1 This figure does not include general correspondence that the Commission received such as emails and letters. This

figure is in relation to explicit replies to the call for evidence.

188


D.4 Summarised below are the responses received according to the themes set out in the call for evidence.

Scheme design

D.5 Whilst some of the responses the Commission received described elements of a new scheme design and how this could operate, most of the evidence instead concentrated on describing fundamental principles which should be enforced through scheme design, such as ‘simplicity’ or ‘adequacy’, or the issues to consider in deciding upon a scheme.

D.6 Of those describing an appropriate scheme, there were a large number of submissions advocating the continued use of final salary schemes. Some of these were prepared to accept changes to other scheme features (such as increases in pension age or contribution rates) if this would make the current schemes sustainable.

D.7 Many other submissions argued for a career average defined benefit scheme as this was perceived as being fair whilst relatively simple to understand. However, there were a number of submissions that advocated a career average scheme only if the benefits were of equal value to those in the current pension schemes. Those submissions which supported career average schemes also stressed the importance of the indexation method used during accrual. Most parties argued that pre-retirement indexation should be based on earnings and not prices. Several submissions cited the recent indexation change from Retail Prices Index (RPI) to Consumer Prices Index (CPI) and that this has led to a devaluation in pensions, especially for the current public service career average scheme, nuvos, where indexation during the accrual phase is now based on CPI. It was felt that for career average schemes, if indexation were based on a measure less generous than earnings, then this should be compensated for with a more generous accrual rate.

D.8 In addition to defined benefit models, several submissions discussed defined contribution schemes, collective defined contribution schemes and how these schemes could be operated in an unfunded model. For example, through the use of notional defined contribution schemes.

D.9 Other schemes that were advocated included hybrids (such as defined benefit schemes with a defined contribution top-up or a career-average/final salary hybrid model), schemes with capped pensionable pay.

Risk-sharing

D.10 There was a wide variety of views from different parties concerning the types and level of risk that members and employers should bear. Some parties argued that members should bear more risk, particularly as the private sector has moved in this direction, whilst others believed that adequate measures were already in place to share further risk, such as current cap and share arrangements.

189


D.11 Suggestions of risk sharing included: the employer bearing the bulk of the risk (in particular investment and inflation), as they are better placed to manage these over the long term; employees sharing the risk of longevity on a progressive basis; and employers bearing investment risks on a core pension with the employee bearing investment risk above this.

D.12 There was a general acceptance amongst most parties submitting to the Commission that longevity was increasing and that this needed addressing. Though a number of submissions stated concern over shifting the increased costs associated with increases in longevity to members, only a small minority of these submissions argued that the employer should bear this risk.

D.13 In terms of contribution rates between the employer and member, there was a large consensus on this ratio being a split of 2:1 employer to employee. Some respondents felt that there should be differences in contribution rates and pension ages for different professions. For example, some respondents felt that a lower pension age should be allowed if physical fitness is an important part of the job, whilst others felt that higher contribution rates should be applied if pension age is lower for certain professions.

D.14 Tiered contributions by income were advocated in some submissions. However, others warned that regard would have to be given to part time staff and to middle income workers who could be ‘squeezed’ to protect lower income groups if tiered contributions were introduced.

Adequacy

D.15 Adequacy of pension income is a subjective judgement and this was reflected in the evidence sent to the Commission. There were a large number of submissions that felt that the post-retirement replacement rates identified by Lord Turner’s Pension Commission were adequate, and others gave figures of either 50 per cent or 67 per cent of pre-retirement income (which would be a full pension earned under either a 1/80ths or 1/60ths final salary scheme).

D.16 Many submissions were of the view that the occupational pension provided by a career in public services, in conjunction with a full basic state pension, should ensure that people have adequate resources in retirement. The reason for this was their view that people generally do not make additional provision, so if the occupational pension in conjunction with a full basic State Pension did not provide an adequate income then more people would end up relying on additional support from the welfare state.

D.17 For people who work part careers in public services, there was a view that benefits should be proportionate to the period of service. However, there was a consensus that arrangements should remain in place to allow people to boost their incomes (by transferring in other benefits, buying added years or contributing to an Additional Voluntary Contributions scheme).

190


Employee understanding and choice

D.18 Most of the submissions suggested that public servants closer to retirement value their pension arrangements more than younger workers, as do those on higher incomes. Equally, the majority of submissions agreed that pension understanding seems generally poor, and so the need for simplicity and strong communication in any new reform programme was emphasised. Submissions on choice were mixed, with some advocating more choice over elements such as contribution rates and retirement ages. Others questioned the concept, arguing that inadequate understanding of pensions would lead to people making poor financial choices if given the option.

D.19 Regarding scheme types most likely to encourage people to save for their retirement, many submissions felt that a defined benefit model would incentivise more people to save, on the premise that more members would be encouraged to join a scheme where risks were shared rather than one in which they bore all the risk.

D.20 Some submissions discussed allowing members to view their retirement savings as a ‘pot’ or cash sum. They argued that this would allow a clearer perception of the value of one’s pension.

Pensions and plurality of provision of public services

D.21 Evidence was very mixed on this issue, yet one element that all parties agreed on was that any reform in this area should seek to make the pensions system more transparent both for the public sector and providers outside of the public sector.

D.22 Parties felt that legislation surrounding the Transfer of Undertakings Protection of Employment regulations (TUPE), the Fair Deal policy and Admitted Body Status was complex, often citing section 75 of the 1995 Pensions Act and the statutory debt this can trigger for employers that cease to participate in the scheme. Many submissions cited the confusion and financial problems this can cause, particularly for third sector organisations.

D.23 There was a mixed response to Fair Deal and the requirement for public service providers to provide ‘broadly comparable’ pensions to transferred workers out of the public sector. Some parties argued that this was necessary in the name of fairness to public service workers that may be outsourced, and others argued that it stifled the innovation of public service providers and contributed to a ‘two-tier’ workforce.2

2 As recommended in the Commission’s interim report the Government launched a consultation on the Fair Deal policy on

3 March 2011.

191


Administration costs

D.24 Most evidence submissions acknowledged the large variation in scheme administration costs, with some suggestions around moving to a shared service model to increase efficiency. There were suggestions around the actual Local Government Pension Scheme (LGPS) Funds themselves, arguing for the combining of Funds to increase their investment efficiency. However others acknowledged that this may also compromise local democracy.

Transition issues

D.25 A large number of submissions advocated the need to consult with unions and scheme governance groups before any reform recommendations are implemented. Appropriate transferring of accrued rights was deemed essential, though there were differing views on whether the final salary link should be maintained or not (views included maintaining the final salary link for future salary increases, treating transferring members as deferred members and breaking the final salary link, or diluting the final salary link to something like CPI plus 1 or 2 per cent).

D.26 Communication and simplicity were also regarded as key elements of any transition. Respondents also noted the need to be aware of any associated administration costs in changing pension schemes.

D.27 Views were mixed regarding who should enter any reformed schemes. Some parties felt that any new arrangement should only apply to new entrants or should be tiered according to time until retirement, whilst others felt that there should be a ‘clean break’ with all staff joining the new scheme.

Local Authority Responses

D.28 A large number of responses came from local councils and local government scheme administrators. There were a number of recurring themes throughout these submissions. This included the need to recognise the LGPS as a different entity to the unfunded public service pension schemes. They felt that the LGPS should not be subject to the same solutions as deemed appropriate for other public service pension schemes. It was generally argued that the LGPS should be retained as a funded scheme, available to a broad range of employers. However there was agreement that mechanisms would need to be further developed to protect the LGPS Funds from employers who cease (for whatever reason) to participate in the scheme and leave any underfunded liabilities.

D.29 It was further argued that Fair Deal should be retained, yet simplified, and the option in the LGPS for contractors to enter into an admission agreement should be retained.3

3 As recommended in the Commission’s interim report the Government launched a consultation on the Fair Deal policy on

3 March 2011.

192


There was opposition towards any move that would consolidate LGPS pension scheme administration and the LGPS Funds, which was thought to be counter to the ‘localism’ agenda.

D.30 It was generally agreed that any reform should seek to deliver a ‘strategic policy framework’ through which individual reforms could take place. There was a strong consensus in favour of a career average model for the LGPS from this group of respondents. However, several councils and overarching bodies emphasised that if, within this model, alternative choices of indexation to earnings were chosen, this should be compensated for by a more generous accrual rate.

Roundtable Events

D.31 Lord Hutton hosted a series of roundtable events during the lifetime of the Commission with various stakeholder groups including unions, employer groups, academics and experts, think tanks, scheme administrators, local authorities, government departments, private sector organisations and third sector organisations. This included events within the devolved administrations.

D.32 Initial roundtable discussions were semi-structured and, much like the first call for evidence, sought to explore the pensions landscape and views amongst interested parties. Roundtable discussions focused on particular issues as the review continued, being structured according to theme and areas that the Commission sought to probe further. Such themes included total remuneration, risk-sharing, adequacy, eligibility, scheme administration, implementation, and the international landscape. These events allowed the Commission to hear the views of stakeholders who were often experts in their area, and their views informed the Commission’s thinking for the final report.

Deliberative Group Event

D.33 On 31 January 2011, the Commission held an all-day deliberative event in London seeking the views and opinions of a sample of public sector workers. There were 89 attendees with a broad diversity in terms of gender and representation across the 6 major public service pension schemes. Slightly more than half of the attendees were aged 46-59, with an under-representation of younger workers and those in part-time and manual positions. It is likely that the attendees had a higher level of pensions knowledge and understanding and expertise than might be typically expected, due to the level of trade union representation at the event and the age profile of the attendees. The event allowed the Commission to gauge the understanding and gain the views of a range of public service workers on pensions and their features. Their opinions informed the Commission’s thinking for the final report.

193


Correspondence

D.34 The Commission received a large volume of correspondence from interested parties during the lifetime of the review, all of which was taken into account in the Commission’s thinking. The Commission received a number of emails relating to pensions contributions made by people to Equitable Life in the past, the difficulty experienced by the pension fund and the role of government throughout this process. As the Commission’s terms of reference relate to public service workers this was not an issue for the Commission to comment on.

Private sector pension schemes data returns

D.35 The Commission sent out questionnaires to approximately 75 UK based private companies requesting information concerning the pension arrangements of the company schemes. Companies were selected on the basis of their size (i.e. number of employees) and where the majority of those employees were based in the UK, to provide a relevant direct comparison with public service pension schemes. The companies chosen reflected a range of employment fields, for example: retail; leisure; banking; construction; insurance; oil; communications; defence; and utilities.

D.36 21 companies responded to the request for information. The questionnaire asked basic questions which included: number of employees; types of schemes operated; opt-out rates; rates of contribution; how decisions are made regarding benefits and contributions; and administrative costs. The answers provided the Commission with opportunities to ask further scheme specific questions, to identify case studies and to invite companies to participate in the roundtable discussions.

194


195


E List of charts, diagrams and tables

Charts

Chart 1.A Period life expectancies for those reaching age 60 – general population

Chart 1.B Projected benefit payments as a percentage of GDP – sensitivity analysis

Chart 1.C Median annual pension receipts for each £100 of contributions

Chart 1.D Membership of employer-sponsored pension schemes among UK employees

Chart 1.E Projected public service pensions expenditure

Chart 2.A Gross replacement rates delivered by typical current public service pension schemes and state pension system

Chart 2.B Gross benefit expenditure by type of member

Box 3.D The distributional impact of moving from final salary to CARE

Chart 3.A Distribution of household wealth excluding pension wealth in 2006-08

Chart 3.B The value of each year of accrued service in a CARE scheme

Chart 3.C Illustrative replacement rates at retirement for a low flyer

Chart 3.D Illustrative replacement rates at retirement for a high flyer

Chart 3.E Proportion of members with careers longer than 20 years achieving adequacy target in a CARE scheme with earnings indexation

Chart 3.F Pension payments from CARE schemes relative to final salary scheme, by deciles of pension income

Charts 3.G Relationship between pension size and male mortality in self-administered pension schemes

Chart 3.H Participation rate by earnings

196


Chart 3.I Percentage of respondents who understand enough about pensions to make a decision about saving for retirement in 2006-08

Chart 4.A Percentage of adult life in retirement, based on NHS Pension Scheme

Chart 4.B Actual and projected period life expectancy at birth for UK males

Chart 4.C Timetable for the changing State Pension Age

Chart 4.D Expected percentage of adult life in retirement after changes

Chart 5.A Distribution of pensionable pay in armed forces, teachers and police schemes

Chart 5.B Distribution of pensionable pay in NHS, civil service and local government schemes

Chart 5.C Variation of career lengths in public service pension schemes

Chart 5.D Projected life expectancy at age 60

Chart 5.E Male life expectancy at age 65 by social class

Chart 5.F The employment status of armed forces leavers

Chart 6.A Economies of scale in pensions administration

Chart C.1 Age effects of schemes based on median earner

Chart C.2 Age effects of schemes based on 90th percentile earner

Chart C.3 Effective employee benefit rates by gender

Chart C.4 Effective employee benefit rates by career progression

Chart C.5 Ratio of benefits received to contributions paid, split by career progression

Chart C.6 Proportion expected to achieve adequacy targets from different scheme designs

Chart C.7 Proportion expected to achieve adequacy targets with at least 20 years’ public sector employment

Chart C.8 Median ratios of pension income to adequacy target from different scheme designs

Chart D.1 Number of submissions received by group

197


Diagrams

Foreword The deal

Box 6.D Proposed governance arrangements for scheme administration

Tables

Table 1.A Recent changes in average effective employee benefit rates for the main schemes

Table 2.A Risks in defined benefit pension schemes

Table 2.B Risks in defined contribution pension schemes

Table 2.C Summary of the Commission’s view of who can bear risks in public service pension schemes

Table 3.A Scheme designs

Table 3.B Risk analysis of different scheme designs

Box 3.B How a CARE scheme could work

Box 3.C How a cash balance scheme with guaranteed conversion terms could work

Table 3.C Results of deliberative workshop survey

Table 3.D Estimated proportion of members affected by a cap on pensionable pay

Table 3.E Assessment of different scheme designs for higher earners against the principles

Table 3.F Scheme designs of equivalent expected average value to members

Table 3.G Current contribution levels

Table 4.A Assessment of longevity management options against the principles

Table 5.A Headline membership information for some of the largest public service pension schemes

Table 6.A Types of governance structures for public service pension schemes

Table A.1 Recommendations

198


Table C.1 PPI proxy scheme structure

Table C.2 Tiered contributions in the PPI proxy scheme

Table C.3 Key financial assumptions

Table C.4 Salary scale assumptions

Table C.5 Scheme designs

Table C.6 Pensions Commission benchmark replacement rates

199


F Glossary

Accrual: A payment earned in one period but not paid until a later period.

Accrual rate: The proportion of earnings that a defined benefit (DB) pension scheme pays as pension for each year of membership. For example, a scheme with an accrual rate of 1/60 provides 1/60th of earnings for each year of membership, which is higher than a pension based on an accrual rate of 1/80th of earnings.

Accrued rights: Rights to pension and other benefits under scheme rules, deriving directly or indirectly from membership of the scheme. Such rights include pension awards already received and pensionable service built up so far based on a particular pension age. However, there is no standard definition of accrued rights across public service pension schemes: the rights will depend on specific circumstances, such as the terms of the individual pension schemes.

Active members: These are current employees who are contributing (or have contributions made on their behalf) to an organisation’s occupational pension scheme. They are distinct from deferred members and pensioners.

Actuarial valuation: A report of the financial position of a DB pension scheme carried out by an actuary every three or four years. The report typically sets out: the scheme’s assets and liabilities as at the date of the valuation; the rate at which the sponsoring employer must contribute to meet the liabilities accruing as they become due; and the additional rate at which the employer must contribute to eradicate any deficit (the excess of liabilities over assets) within a stated time period.

Additional Voluntary Contribution (AVC): These are personal pension contributions made by someone who is also a member of an occupational scheme as a top-up to their occupational entitlement. AVCs can be made into the occupational scheme or to a stand-alone product called a Freestanding AVC plan.

Admitted Body Status: Admitted body status refers to the practice of the Local Government Pension Scheme of accepting as members the employees of bodies not covered by the original or primary ambit of the scheme as set out in its founding statute. It enables contractors, who take on an authority’s services or functions with employees transferring from the authority, to offer the transferring staff continued eligibility of the transferring authority’s pension scheme.

Ancillary benefits: Additional benefits not directly linked to the pension itself. These include death benefits, ill-health benefits, dependants’ benefits and pension guarantees.

200


Annuity: A series of regular payments usually payable for the life of the annuitant. Annuities are usually purchased by a lump sum of cash. Pension schemes sometimes discharge their promise of pension benefit by purchasing an annuity. Individuals can purchase an annuity using their own capital. There is a wide range of options available e.g. level, escalating, guaranteed, single or joint lives.

Automatic enrolment: A pension scheme where an individual is made a member by default and actively has to decide to leave the scheme.

basic State Pension (bSP): Non-earnings-related pension based on an individual’s National Insurance Contribution record.

Cap and share: This is an arrangement applying to the pension schemes for the NHS, Teachers, Civil Service and Local Government, whereby increases or reductions in the costs of a scheme identified in a pension scheme actuarial valuation are shared between employees and employers up to the value of the cap. Above the cap the increases or reductions are borne by employees, either by changing employee contributions or the cost of employee benefits or both. Below the cap, increases or reductions are shared between employers and employees.

Capped scheme: A pension scheme where a limit is placed on pension entitlement, for example by placing a ceiling on the amount of annual earnings that are pensionable or by limiting the amount of pension that might be awarded under scheme rules.

Career average scheme: A defined benefit scheme that gives individuals a pension based on a percentage of the salary earned in each year of their working life.

Cash balance scheme: A scheme where the employer puts a notional amount into the member’s pension pot every year, which is then guaranteed. This credit can be expressed as a percentage of salary for each year worked. If cash contributions from the employee and employer, plus investment returns, do not match this promised notional credit then the employer has to meet any shortfall. On retirement the resulting cash balance can be converted into an annual income stream.

Cohort Life Expectancy: The estimate of an individual’s probability of surviving future years allowing for changes in mortality rates over time.

Collective DC scheme: All member pension contributions are placed in one fund that is then managed on behalf of the members. As in standard DC schemes the pensions will vary according to the value of the underlying investments. However, within collective DC schemes there is the option to spread the effects across the various groups of members (intergenerational sharing) to smooth the effects of market conditions.

Commission: The Independent Public Service Pensions Commission.

201


Commutation factor: A number used to convert a pension annuity into a lump sum. The factor usually depends on the sex of the member and the age at which the conversion takes place. The factors are scheme specific and are either set out in the pension scheme’s rules or are updated periodically by the scheme’s trustees or administrators.

Conditional indexation: Where the uprating of a pension fund or pensions in payment each year is variable and dependent on other factors, such as investment returns.

Consumer Prices Index (CPI): An internationally comparable measure of inflation based on structures in international legislation and guidelines and launched in 1996. Like the Retail Prices Index (RPI) it tracks the changing cost of a fixed basket of goods and services over time. However unlike the RPI it disregards some items, such as housing costs. It also has a different population base for the indices from the RPI and a different way in which the index is calculated.

Contracting-out: The facility to opt out of the additional state pension and build up benefits in a private pension scheme.

Conversion factor: A ratio which determines how much money is needed in the pension pot to purchase £1 of annual pension income.

Cost sharing: The cost of any benefit increases is shared between individual and employer.

Current contribution rate: The standard contribution rate as adjusted for past surpluses and deficits and payable by employers and employees

Current service cost: A measure of the value of the new pension promises built up over a year.

Death benefits: A pension scheme benefit that is usually paid to the dependant of a scheme member if that member dies. Death in retirement benefits typically take the form of a pension paid to the dependant of a proportion of the pension the member was receiving when he or she died. Death in service benefits typically take the form of a lump sum (Death Benefit Lump Sum), calculated as a multiple of salary, plus a pension paid to the dependant of a proportion of the pension the member would have received if he or she had lived until retirement age.

Deferred members: Deferred members are scheme members who have left employment, or ceased to be an active member of the scheme whilst remaining in employment, but retain an entitlement to a pension from the scheme.

Deferred pension: A pension that will be payable to a deferred member when he or she chooses to draw it.

Defined benefit (DB) pension scheme: A pension scheme where the pension is related to the members’ salary or some other value fixed in advance.

202


Defined contribution (DC) pension scheme: A scheme where the individual receives a pension based on the contributions made and the investment return that they have produced. These are sometimes referred to as money purchase schemes.

Dependant member: An individual who is eligible to receive retirement benefits following the death of a scheme member.

Employee contribution rates: The percentage of their pensionable salary that employees pay as a contribution towards a pension.

Employer contribution rates: The percentage of the salary of employees that employers pay as a contribution towards the employees’ pension.

Fair Deal: A non-statutory code of practice introduced in 1999 that protects the pension provisions of public sector workers who have their employment compulsorily transferred out of the public sector. In such a situation the transferring organisation is required to ensure that the pension provision for future service is broadly comparable after the transfer.

Final salary scheme: A DB scheme that gives individuals a pension based on the number of years of pensionable service, the accrual rate and final earnings as defined by the scheme.

Funded: Pension schemes in which pension contributions are paid into a fund that is invested and pensions are paid out of this pot.

Hybrid scheme: A scheme which incorporates both defined benefit and defined contribution elements of benefit provision.

Independent Public Service Pensions Commission: An independent commission undertaking a fundamental structural review of public service pension provision by Budget 2011

Indexation: The technique used to adjust income payments or the uprating of a pension fund in line with an index.

Life expectancy: Life expectancy at a given age, x, is the average number of years that a male or female aged x will live thereafter.

Longevity: The length or duration of human life.

Member contributions: The amounts paid by active scheme members into their pension schemes.

Mutualisation: Employee participation in, and of, an organisation. Implied sharing of the risks and benefits.

203


National Insurance (NI): The national system of benefits paid in specific situations, such as retirement, based on compulsory earnings-related contributions by employers and employees. Self-employed people make contributions on a different basis.

NEST (National Employment Savings Trust): The arms length from Government, low cost pensions scheme associated with the automatic enrolment reforms planned for 2012.

Net cash expenditure: Benefits paid to recipients less contributions received by central government from employees and employers in one year.

Normal Pension Age: The earliest age at which, in the normal course of events, a scheme member may retire with payment of his or her unreduced accrued superannuation benefits.

Notional defined contribution scheme: A scheme whereby the values of the pensions at retirement are determined by an assumed return on contributions and an annuity rate or rates.

Occupational pension: A pension, which is provided via the employer, but from a pension scheme that typically takes the form of a trust arrangement and is legally separate from the employer.

Open market annuity: An annuity purchased from the competitive insurance market.

Pension Credit: The main income-related benefit for pensioners, which combines the Guarantee Credit and the Savings Credit.

Pensioner member: Individuals who now draw a pension and who are mainly former employees. However they may also include widows, widowers and other dependants of former active members.

Period Life Expectancy: Represents the amount of time an individual is expected to live if mortality rates were equal to the experience of other individuals in that year.

Public sector pension schemes: These comprise both public service pension schemes and other schemes in the wider public sector such as the BBC, Transport for London, the Bank of England and the Royal Mail. These schemes are not authorised by statute and the organisation concerned makes the rules of the schemes.

Public Sector Transfer Club: A group of some 120 salary related occupational pension schemes. It allows easier movement of staff mainly within the public sector. It does this by making sure that employees receive broadly equivalent credits when they transfer their pensionable service to their new scheme regardless of any increase in salary when they move to their new employment.

Public service pension schemes: Pension schemes authorised by statute where the relevant Ministers make the rules of the schemes. The main schemes are those for civil servants,

204


the armed forces, NHS employees, teachers, local government employees, the police and firefighters. There are over 200 public service pension schemes.

Replacement rate: The ratio of pension income to salary at retirement.

Retail Prices Index (RPI): A measure of inflation and like the Consumer Prices Index (CPI) it tracks the changing cost of a fixed basket of goods and services over time. However, unlike the CPI it takes into account items such as housing costs. It also has a different population base for the indices from the CPI and a different way in which the index is calculated.

SCAPE (Superannuation Contributions Adjusted for Past Experience): A methodology used to set employer contribution rates across public service intended to mirror the operation of a funded scheme by keeping track of a notional ‘Pension Account’.

Scheme liabilities: The scheme liabilities at a given date are an estimate of the total value of future payments that the scheme will have to make to all scheme members in respect of pension rights which have been earned before that date.

State Pension Age (SPA): The age from which an individual can claim their state pension. It is currently 65 for men and will increase to 65 for women by November 2018.

State Second Pension (S2P): The National Insurance pension that gives benefits based on an individual’s earnings and contributions.

Top-up DC: Where a DC arrangement is available to supplement another form of pension provided by an employer.

TUPE: Transfer of Undertakings (Protection of Employment) Regulations 2006.

Unfunded pension schemes: Pension schemes, which are not backed by a pension fund. Instead current contributions are used to pay current pensions along with other funds provided by the employer. Most public service schemes are unfunded, except for the Local Government scheme, which is funded.

205


G Abbreviations

CARE Career Average Revalued Earnings

CPI Consumer Prices Index

DB Defined Benefit

DC Defined Contribution

DCLG Department for Communities and Local Government

DWP Department for Work and Pensions

ESA European System of Accounts

GAD Government Actuary’s Department

GDP Gross Domestic Product

GP General Practitioner

HMRC Her Majesty’s Revenue and Customs

IPSPC Independent Public Service Pensions Commission

IT Information Technology

LGPS Local Government Pension Scheme

LPFA London Pensions Fund Authority

NDPB Non-departmental public body

NEST National Employment Savings Trust

NHS National Health Service

NHSPS National Health Service Pension Scheme

NPA Normal Pension Age

206


OBR Office for Budget Responsibility

OECD Organisation for Economic Co-operation and Development

ONS Office for National Statistics

PCSPS Principal Civil Service Pension Scheme

PPI Pensions Policy Institute

RPI Retail Prices Index

SCAPE Superannuation Contributions Adjusted for Past Experience

SPA State Pension Age

TPS Teachers Pension Scheme

207


H Bibliography

Audit Commission (2010), Local government pensions in England: An information paper

Audit Scotland (2011), The cost of public sector pensions in Scotland

Barr, N. and Diamond, P. (2008), Reforming Pensions

Barr, N. and Diamond, P. (2009), ‘Reforming pensions: principles, analytical errors and policy directions,’ International Social Security Review, vol. 62, issue 2, pp.5-29

Clery, E., Humphrey, A. and Bourne, T. (2009), ‘Attitudes to pensions: the 2009 survey,’ Department for Work and Pensions Research Report no. 701

Continuous Mortality Investigation (2008), Working Paper 31

Department for Communities and Local Government (2009), Admitted body status provisions in the Local Government Pension Scheme

Department for Communities and Local Government (2010), Local Government Financial Statistics England, no. 20

Department for Communities and Local Government (2010), Fire and Rescue Service: Operational Statistics Bulletin for England 2009-10

Department for Work and Pensions (2009), Attitudes to Pensions: the 2009 survey

Department for Work and Pensions (2010), A sustainable State Pension: when the State Pension age will increase to 66

Department for Work and Pensions (2010), Older Workers Statistical Information Booklet, Quarter Two 2010

Emmerson, C., Reed, H. and Shephard, A. (2004), ‘An assessment of Pensim2,’ IFS Working Paper no. 04/21

Feldstein, M. and Liebman, J. (2002), ‘Social Security,’ in Handbook of Public Economics, ed. A. Auerbach and M. Feldstein

HM Treasury (2000), Review of Ill-Health Retirement in the Public Sector

http://www.audit-commission.gov.uk/SiteCollectionDocuments/AuditCommissionReports/NationalStudies/100729localgovernmentpensions.pdf



208


HM Treasury (2010), Spending Review 2010

Hutton Review of Fair Pay (2010), Interim Report

Independent Public Service Pensions Commission (2010), Interim Report

Institute for Fiscal Studies (2009), The IFS Green Budget January 2009

Institute for Fiscal Studies (2011), The IFS Green Budget February 2011

Ippolito, R. (2002), ‘Stayers as ‘Workers’ and ‘Savers’: Toward Reconciling the Pension-Quit Literature,’ Journal of Human Resources, vol.37, no.2, pp.275-308

Iyengar, S. and Kamenica, E. (2010), ‘Choice proliferation, simplicity seeking, and asset allocation,’ Journal of Public Economics, vol. 94, issues 7-8, pp.530-539

Laibson, D. (1996), ‘Hyperbolic Discount Functions, Undersaving, and Savings Policy,’ NBER working paper no.5635

Marmot Review (2010), Fair Society, Healthy Lives

Matthews, D. (2010), ‘The changing face of public sector employment 1999-2009,’ Economic and Labour Market Review, July

Modigliani, F. and Brumberg, R. (1954), ‘Utility Analysis and The Consumption Function: An Interpretation of Cross-Section Data,’ in Post-Keynesian Economics, ed. K. Kurihara

National Association of Pension Funds (2009), Annual Survey 2009

National Audit Office (2007), Leaving the services

National Audit Office (2010), The cost of public service pensions

National Audit Office (2010), The impact of the 2007-08 changes to public service pensions

Office for Budget Responsibility (2010), Economic and fiscal outlook, November

Office for National Statistics (2010), Occupational Pension Schemes Annual Report 2009

Pensions Commission (2004), Pensions: Challenges and Choices – The First Report of the Pensions Commission

Pensions Commission (2005), A New Pension Settlement for the Twenty-First Century - The Second Report of the Pensions Commission

http://www.nao.org.uk/publications/0910/public_service_pensions.aspx

209


Pensions Policy Institute (2008), An assessment of the Government’s reforms to public sector pensions

Pensions Policy Institute (2010), Public sector pension schemes: policy objectives and options for the future

Pensions Policy Institute (2010), The future of the public sector pensions

The Pensions Regulator (2009), Scheme funding: An analysis of recovery plans

PricewaterhouseCoopers (2011), Public service pensions reform: findings from consumer research

Railway Pensions Commission (2008), Final Report

Shaw, C. (2007), ‘Fifty years of United Kingdom national population projections: how accurate have they been?’ Population Trends, no.128, Office for National Statistics

http://www.pensionspolicyinstitute.org.uk/uploadeddocuments/PPI_public_sector_pensions_16_Oct_2008.pdf

http://www.pensionspolicyinstitute.org.uk/uploadeddocuments/PPI_public_sector_pensions_16_Oct_2008.pdf

https://www.pensionspolicyinstitute.org.uk/uploadeddocuments/2010/PPI_Discussion_Paper_on_PSP_12_March_2010.pdf

https://www.pensionspolicyinstitute.org.uk/uploadeddocuments/2010/PPI_Discussion_Paper_on_PSP_12_March_2010.pdf

http://www.thepensionsregulator.gov.uk/docs/scheme-funding-analysis-2009.pdf

210


Independent Public Service Pensions Commission: Final Report

Documents

Transcript of Independent Public Service Pensions Commission: Final Report